INSIGHTS OF CONTEMPORARY DIFFUSION WEIGHTED IMAGING SIGNAL 
MODELING TECHNIQUES ON WHITE MATTER MICROSTRUCTURAL CHANGES 
FOLLOWING MILD TRAUMATIC BRAIN INJURY  

By 

Joshua H. Baker 

A DISSERTATION 

Submitted to  
Michigan State University  
in partial fulfillment of the requirements  
for the degree of  

Neuroscience-Doctor of Philosophy 

2024 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ABSTRACT 

Mild traumatic brain injury, which accounts for up to 90% of traumatic brain injuries, is 

currently diagnosed and monitored with a thorough history and physical exam. There is growing 

consensus in the literature that pathophysiologic changes in white matter extend past symptomatic 

recovery, and that these biological changes should be the target of diagnosis and monitoring. This 

would replace the current clinical consensus that symptom resolution and a level of functioning that 

allows return to work/play indicates recovery. Diffusion magnetic resonance imaging has been a 

critical tool for studying white matter change, and the diffusion tensor model developed in the 1990s 

has been the cornerstone of studying white matter in human subjects with mild traumatic brain 

injury. This model has several limitations, however, which in part may contribute to inconsistencies 

observed in the localization and direction of changes in diffusion tensor metrics such as fractional 

anisotropy. The past decade has brought forth scanner advancements, including stronger gradients, 

high angular-resolution imaging, and modeling strategies specific to these imaging data. Here, we 

review these contemporary techniques and apply them in a secondary analysis of data collected 

during the Transforming Research and Clinical Knowledge in Traumatic Brain Injury Study. 

Constrained spherical deconvolution and fixed-based analysis were applied to diffusion magnetic 

resonance imaging data collected during the study, revealing significant differences in tract-specific 

and global fiber-density and fiber-density cross-section measures between patient and control 

groups. However, results remained relatively stable over time, indicating that in this population-

based sample of patients, not much change occurred in the white matter from 2 weeks to 6-months 

post-injury. This would seem to suggest there is a drastically different timeline for white matter 

recovery than was previously thought, even in injuries categorized as mild in the general population. 

Additionally, this reinforces the notion that multimodal tools for diagnosis and monitoring need 

further evaluation to create gold-standard objective classifiers of injury severity. In the short-term, 

 
 
 
future work should focus on the development of an imaging technique that when collected in the 

acute period following injury can predict prolonged recovery in individual patients with reasonable 

accuracy. Finally, characteristics which convey resiliency to poor outcomes should be investigated to 

facilitate development of interventions that hasten biological recovery.

 
This dissertation is dedicated to my loving family.  
Mom & Dad thank you for your sacrifice and love. 
Mimi, Granduke & Gary, this wouldn’t have been possible. 
without your support.  
Finally, to my partner Sarah. Because of you, I live.  
I’ll always be by your side, loving you.  

  iv 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ACKNOWLEDGEMENTS  

First, I want to acknowledge the DO/PhD program that has funded my work and 

education. Also, Drs. Brian Schutte and John Goudreau who helped me form ideas and think 

critically as a scientist. Of course, endless thanks to the amazing Michelle Volker she took care of so 

many things behind the scenes so I could focus on the science. Additionally, I would like to thank 

Dr. AJ Robison and Eleri Thomas for their guidance and cheerful attitudes in the Neuroscience 

program. I want to acknowledge my lab mates and colleagues Dr. Zac Fernandez and Joshua Hubert 

for always being there with the boost I needed.  Your collaboration and support on projects helped 

me grow into the scientist I am today. When I first came to MSU, I was set to work on a high school 

concussion study for my dissertation, and would like to thank the MSU/Sparrow Concussion clinic 

staff especially Jill Leavitt who coordinated recruitment from the side of the concussion clinic and 

Dr. Matthew Saffarian who trained me to administer the SCAT5. I want to acknowledge Dr. Kan 

Ding, who provided feedback on the TRACK-TBI proposal. The TRACK-TBI investigators are the 

only reason the data for this project exists and allowed me to reap the benefits of their insight into 

addressing the great need for open TBI research; a special thank you to Dr. Pratik Mukhurjee. 

Thank you to the employees, administrators, and scientists at FITBIR who facilitated the simple 

acquisition of the data from their revolutionary repository. Finally, I am eternally grateful to the 

members of my guidance committee who mentored me, listened to my ideas and gave invaluable 

feedback to improve me work. I want to specifically thank Dr. Rebecca Knickmeyer who was always 

available for discussions, and gave feedback regularly on the work at poster presentations and other 

neuroscience events during my training. Also, Dr. Norman Scheel, the world’s most metal 

computational neuroscientist, made this project possible with his insights and as my computer 

programming mentor. A profound thank you to Dr. Andrew Bender, who has been a kind, 

nurturing mentor and challenged me to learn and do more. His expertise in diffusion, and as an 

  v 

endless source of MRTrix3 metadata was invaluable during the preparation of this document. 

Finally, I would like to thank Dr. David Zhu, who gave me the opportunity to start down a new 

path of discovery in my research career in his laboratory and taught me everything I know about 

magnetic resonance imaging. 

  vi 

 
 
PREFACE 

The field of neuroimaging is unique in that it requires a working base of knowledge in MRI 

physics, computer science, statistics, and software development, in addition to expertise in the 

disease area in which you study. This massive amount of information is difficult to maintain in a 

single person, and all fields are rapidly advancing with novelty giving way to antiquity with singular 

publications. Neuroimaging, therefore, is a highly collaborative team-based science discipline. It is 

easy to overestimate one’s knowledge in any one of the areas as a neuroimaging researcher, and 

blind spots are abundant in the literature including this dissertation. Therefore, a critical piece of 

advice for the neuroimaging practitioner is a piece of wisdom from Clint Eastwood’s character 

“Dirty” Harry Callahan in Magnum Force, “A mans got to know his limitations”. Obviously, this is 

partially in jest. However, this phrase could be updated for more inclusive modern times and made 

specific to neuroimaging with the phrase “for a person using neuroimaging in research knowing 

what you don’t know is almost more important than what you do, consult colleagues with more 

knowledge in these intersecting fields often.”  

With the recent advances and decreasing cost of computers, this incredible field has become 

accessible to more individuals than ever. A challenge in the neuroimaging of mild traumatic brain 

injury is the ever-changing set of tools and evolving technologies with which the disease is studied. 

Nonetheless, diffusion MRI and other quantitative imaging techniques have been a mainstay in 

human subjects’ research on this traumatic event for decades. During the 21st century, these 

neuroimaging research tools will likely translate from keyboard to bedside and be used in numerous 

applications to classify historically ambiguous neuropsychiatric diseases with the precision of in vivo 

imaging. Indeed, a major area of research will focus on delineating the numerous factors that drive 

resiliency to injury and that underlie various phenotypes of recovery in the general population. 

Devising model systems to test novel therapeutics to intervene in cases of poor recovery will be 

 vii 

informed by ongoing large population-based studies. It is critical in my opinion to first fully describe 

the phenomenology of brain trauma in humans. Much of this hinges on the ability to localize and 

characterize changes in the brain accurately and predict who is likely to have a poor outcome. This 

will allow researchers to focus on these individuals for a more detailed understanding of what drives 

this phenomenon. Quantitative MRI is likely one of the best tools we currently have available to 

accomplish this. As a physician-scientist in training aspiring towards a career in Diagnostic 

Radiology, the knowledge, and skills developed in the preparation of this dissertation could be 

applied widely in numerous disease processes whose pathophysiology affect white matter. In 

addition, much of our knowledge on the embryology of the developing human brain come from 

animal studies, and advanced imaging techniques hold promise to unlock the intricate processes of 

the developing human brain by scanning the fetus while still in the womb. Finally, as a classically 

trained anatomist, my perspective is that structural brain imaging is the modern scalpel with which 

advances in our understanding of white matter morphology will continue to evolve. Presently, the 

anatomical record of white matter is a wild west of tracts and fibers organized by function, location, 

and cell type. As a classically trained anatomist with expertise in structural neuroimaging, I hope to 

contribute to classification of white matter and add to our understanding of the relationship between 

white matter lesion locations and brain function during my career. Needless to say, there is much 

work yet to be done, or as the great character Sherlock Holmes from the writings of Sir Arthur 

Conan Doyle said, “The game is afoot”.  

 viii 

 
 
TABLE OF CONTENTS 

CHAPTER 1: INTRODUCTION ................................................................................................................. 1 
MILD TRAUMATIC BRAIN INJURY ................................................................................................... 4 
DIFFUSION MAGNETIC RESONANCE IMAGING .................................................................... 14 
DIFFUSION TENSOR IMAGING IN MILD TRAUMATIC BRAIN INJURY ......................... 21 

CHAPTER 2: A RAPID REVIEW OF MILD TRAUMATIC BRAIN INJURY STUDIES 
UTILIZING CONTEMPORARY TECHNIQUES FOR DIFFUSION IMAGING      
ANALYSIS ........................................................................................................................................................ 29 
INTRODUCTION ..................................................................................................................................... 29 
HIGHER ORDER SIGNAL REPRESENTATION AND BIOPHYSICAL MODELLING 
TECHNIQUES ........................................................................................................................................... 30 
MATERIALS & METHODS ................................................................................................................... 36 
RESULTS & DISCUSSION ..................................................................................................................... 37 

CHAPTER 3: FIXEL-BASED ANALYSIS PIPELINE TESTING IN A TRACK-TBI 
SUBSAMPLE .................................................................................................................................................... 78 
INTRODUCTION ..................................................................................................................................... 78 
MATERIALS & METHODS ................................................................................................................... 79 
RESULTS ...................................................................................................................................................... 84 

CHAPTER 4: A POPULATION-BASED SAMPLE OF MILD TRAUMATIC BRAIN INJURY 
REVEALS ABNORMAL WHITE MATTER STRUCTURE 6-MONTHS                             
POST-INJURY WITH FIXEL-BASED ANALYSIS: A TRACK-TBI STUDY ................................. 86 
INTRODUCTION ..................................................................................................................................... 86 
MATERIALS & METHODS ................................................................................................................... 88 
RESULTS ...................................................................................................................................................... 96 

CHAPTER 5: FIXEL-BASED ANALYSIS IN A SAMPLE OF HIGH SCHOOL ATHLETES 
WITH SPORTS-RELATED MILD TRAUMATIC BRAIN INJURY ............................................... 111 
INTRODUCTION ................................................................................................................................... 111 
MATERIALS & METHODS ................................................................................................................. 113 
RESULTS .................................................................................................................................................... 115 

CHAPTER 6: GENERAL DISCUSSION & FUTURE DIRECTIONS............................................ 118 
GENERAL DISCUSSION OF EXPERIMENTAL CHAPTERS .................................................. 118 
FUTURE DIRECTIONS ........................................................................................................................ 127 
CONCLUSION ......................................................................................................................................... 132 

REFERENCES .............................................................................................................................................. 133 

APPENDIX .................................................................................................................................................... 153 

  ix 

 
 
 
 
 
 
 
 
 
 
CHAPTER 1: INTRODUCTION 

Mild traumatic brain injury (mTBI) is a classification currently used by clinicians and 

researchers to describe a limited range of brain injuries resulting from a traumatic event. Currently, 

designating a traumatic brain injury as “mild” often refers to injuries causing heterogeneous clinical 

sequelae considered to be modest and that typically resolve entirely within weeks.1 However, the lack 

of prognostic value of the current nosology of traumatic brain injury poses a major challenge for 

clinicians and scientists. In the current framework, categorizing an injury as ‘mild’ has little clinical 

significance and fails to predict the clinical course of recovery; therefore, the mTBI classification 

provides clinicians little guidance regarding patient counseling and treatment options. Notably, there 

is currently no treatment targeting the biological underpinnings of mTBI.2 In addition, despite 

considerable extant research efforts, there are neither reliable objective prognostic biomarkers, nor a 

clear understanding of the dynamic brain changes occurring during recovery.3,4 Early studies often 

focused on predominantly white male populations engaged in contact sports or military personnel, 

limiting the generalizability of the findings to date.5 One promising source of potential diagnostic 

and prognostic biomarkers is in vivo magnetic resonance imaging (MRI), including diffusion 

magnetic resonance imaging (DWI).3,17 Neuroimaging methods sensitive to neurophysiologic 

changes from the traumatic injury could provide new valuable prognostic approaches for these 

patients.  

Currently, there is no validated prognostic biomarker for patients with brain injuries 

classified as mild. Recent studies report that up to 50% of mTBI patients report persisting 

symptoms at one-year post-injury that are significantly worse than baseline pre-injury symptoms. 6–8 

This length of symptom persistence represents a significant divergence from current frameworks 

which underestimate the time course and severity of prolonged symptoms following mTBI.9 Clinical 

  1 

 
 
 
 
trials to develop targeted treatments for individuals who develop prolonged symptoms will be 

challenging without a prognostic biomarker, as enrollment weighting to capture those who are at the 

most significant risk of developing prolonged symptoms cannot be accomplished without one. 

Several potential candidates exist, including various neuroimaging techniques10,11, serum 

biomarkers12,13, and clinical assessment scores. Combining these measures in a multivariate 

prognostic model will likely have the highest accuracy due to the dynamic and heterogeneous nature 

of this injury.14 

Evidence from neuroimaging studies have identified differences in white matter (WM) and 

brain function that occur during mTBI recovery. Neuroimaging studies have correlated observed 

changes during recovery with poor outcomes on the Glasgow outcome scale-extended (GOSE) and 

Rivermead post-concussion symptoms questionnaire (RPQ).15 DWI, specifically, assesses the 

diffusion/random  movement of water molecules to infer changes in WM microanatomy indirectly. 

Diffusion Tensor Imaging (DTI) is a modeling technique for estimating parameters such as 

Fractional Anisotropy (FA), an inference of microstructural coherence of WM fibers, which is often 

considered a proxy for WM integrity and the capacity to carry signals between brain regions.16 An 

early examination of the Transforming Research and Clinical Knowledge in Traumatic Brain Injury 

(TRACK-TBI) Study demonstrated that DTI regions of interest (ROI) with FA 2.2 standard 

deviations below controls were predictive of poor recovery, having >2-3 symptoms worse than pre-

injury, at 3- and 6-month follow-up.15 Several systematic reviews of DWI studies in mTBI add 

further strong evidence that these measures can identify differences that current clinical MRI 

sequences and CT cannot.3,5,17 

DWI studies of the acute and subacute phases following injury, defined as the first 7 days 

and between 1 week and 3 months following injury, respectively, demonstrated a pattern of reduced 

FA and increased mean diffusivity (MD)/radial diffusivity (RD) in frontal and deep white matter 

  2 

 
regions.3,17 It is postulated that this reflects demyelination and inflammation related to diffuse axonal 

injury. However, in this same review, conflicting patterns of FA and MD were consistently 

observed, casting uncertainty onto the true pattern of DWI metric change following mTBI and how 

to interpret it. More consistent injury to scan time, acquisition parameters and modeling techniques 

will help translate this body of literature into clinical practice. However, there are significant 

limitations of DTI that cannot be ignored, which will be covered in this dissertation. Therefore, 

analysis with novel and complimentary DWI modeling techniques may reveal more about the 

pattern of post-injury brain changes with greater interpretability regarding the underlying 

pathophysiology observed following a mTBI. 

This dissertation details our application of constrained spherical deconvolution (CSD), a 

higher-order DWI modeling technique developed by the creators of the diffusion analysis software 

package MRTrix3.18 We concatenate two separate DWI sequences with b-values of 1300 mm/s2 and 

3000 mm/s2 collected during the same scanning session to create a pseudo-multi-shell dataset. Using 

this process, we used legacy data for a more specific secondary analysis of the diffusion data from 

the open-access Federal Interagency Traumatic Brain Injury Repository (FITBIR). We then 

compared control and mTBI patients using an analysis method developed for MRTrix3 termed 

“fixed-based analysis (FBA).”19 As will be subsequently expanded upon, a ‘fixel’ refers to a specific 

fiber population within a voxel. Unlike traditional voxel-based techniques like DTI, FBA defines a 

novel volume element for inference with DWI data and utilizes a more biologically plausible 

modeling technique with potentially greater specificity for underlying WM microstructural change.20 

A fixel can represent more than a single distinct fiber bundle within a voxel, allowing inference of 

effects beyond simply within voxel microstructure. FBA addresses some of the significant 

limitations of the DTI model and analysis with Tract-Based Spatial Statistics (TBSS) by addressing 

  3 

 
the crossing fibers issue and allowing comprehensive analysis of WM tracts by addressing partial 

volume effects at the interface of tissue types.20–22 

This exciting area of research falls at the intersection of many rapidly changing fields, 

including the clinical research of traumatic brain injury, the advancement of DWI pulse sequences, 

and the development of new neuroimaging modeling and analysis techniques.  

MILD TRAUMATIC BRAIN INJURY 

A concussion, or interchangeably mTBI, is damage to the brain caused by a transfer of 

mechanical force to neuronal tissue 23,24. The molecular and cellular changes secondary to this impact 

result in a metabolic deficit, inflammation, and damage to the blood-brain barrier (BBB), which are 

incompatible with normal brain homeostasis. Here, the current knowledge of the epidemiology, 

pathophysiology, and management of mTBI will be discussed. Additionally, there will be a 

translational emphasis on studies that attempt to link cellular and molecular changes in the brain to 

the diverse symptoms of this injury. 

The epidemiology of traumatic brain injury, especially those that are categorized as mild, is a 

challenging area of literature to summarize. This is mainly due to the historical heterogeneity of 

classification and siloed consensus statements from specific professional societies that were slow to 

gain exposure in the mainstream research community.25 This is further complicated by the two areas 

of literature, concussion, and mTBI, being in largely separate disciplines, which likely contributed to 

the slow adoption of this classification.26 For example, concussion research, especially with a sports 

etiology, has generally been housed within fields such as sports medicine and kinesiology. In 

contrast, mild traumatic brain injury work is often conducted by groups composed of emergency 

medicine, neurology, psychiatry, and neurosurgery researchers.  

Mild TBI represents the dominant type of TBI, comprising 70-90% of injuries and creating an 

estimated worldwide financial burden of $400 billion annually.9,27 In the US, the CDC reported 

  4 

 
 
traumatic brain injury accounted for 2.5 million emergency department visits in 2010.28 Epidemiologic 

studies have demonstrated a roughly 2:1 ratio of male to female traumatic brain injuries when 

including all severity levels; however, it is important to note that these studies were conducted on data 

collected before 2015. At the time of writing, more recent nationwide estimates are not available. The 

primary precipitating events in mTBI vary by age, with motor vehicle accidents being the leading cause 

in younger individuals and falls being most common in older individuals. In pediatric populations, the 

most common mechanism of mild TBI is falls (~50%), followed by impact injuries such as patient vs 

vehicle (~25%).29 Importantly, 30% of individuals over the age of 65 fall each year, and since 1999, the 

incidence of death from falls has more than doubled.30 It is likely that these statistics foreshadow a 

similar increase in traumatic brain injury morbidity and mortality among the aging US population, with 

fall prevention a major potential mechanism for healthcare professionals to intervene.  

During a traumatic impact, mechanical forces are transferred to neural tissue, which are 

hypothesized to transiently disrupt the lipid bilayer by a mechanism called mechanoporation.1 

Ultimately, mechanoporation leads to the precipitating events for the pathophysiologic cascade of 

mTBI. It causes an efflux of potassium exacerbated in part by the mechanical opening of voltage-

gated potassium channels and also a subsequent opening of ligand-gated ion channels after the 

indiscriminate release of glutamate 31,32. As a result, a cascade of events is triggered by the opening of 

ligand-gated ion channels and the cells’ attempt to restore ionic equilibrium. Potassium flux and 

release of glutamate return to baseline levels within 24 hours of injury 33. Despite this, a series of 

events triggered by the imbalance of ions plague the cells for days, weeks, and sometimes months 

post-injury. Most of our knowledge of the timeline of these pathophysiologic processes is 

extrapolated from animal studies or early human studies utilizing serum biomarkers with small 

sample sizes. 

  5 

To restore ionic equilibrium, there is increased activity of the sodium-potassium ATPase.34 

Animal models have shown that increased ATPase activity is reflected in a biphasic change within 

cerebral metabolism. In the hours following the mTBI injury, there is a significant increase in 

glucose metabolism followed by a depression which lasts up to 10 days in adult rats 35,36. The 

depression in glucose metabolism is linked in part to mitochondrial dysfunction.37 Another 

significant contributor to the pathophysiology of mTBI is calcium flux into the neuron cell body, 

contributing to mitochondrial dysfunction. 

Neuronal calcium concentration is maintained at low concentrations compared to 

extracellular levels due to its cytotoxic effects. As a result of the indiscriminate release of glutamate 

after impact, there is an acute increased activation of the N-methyl-D-aspartate receptor (NMDAR) 

through which calcium enters the cell. Interestingly, it has been shown that molecular changes to the 

NMDAR occur following mTBI, and that these changes directly affect the conduction of calcium 

across the neuronal membrane. As a result, following mTBI, there is increased sensitivity to 

glutamatergic activity and increased intracellular calcium concentrations.38,39 These changes to the 

NMDAR have been shown to persist for as long as 10 days following injury in animal models.37 

Neurons can buffer calcium with intracellular binding proteins and sequestration in organelles. 

Mitochondria also play a role in sequestering calcium; however, it results in the dissipation of the 

chemical gradient necessary for energy production via the electron transport chain.40  The burden of 

calcium sequestration on mitochondria further exacerbates the mismatch between energy demands 

after mTBI and energy production. 

In mTBI, mitochondria in damaged neurons ultimately fail to sequester calcium. This leads 

to increased intracellular calcium concentration and mitochondrial dysfunction, bringing into focus 

an increasing energy deficit for damaged neuronal cells. This situation is made worse by a mismatch 

in cerebral blood flow (CBF) and metabolic demand following mTBI injury, termed flow-

  6 

metabolism uncoupling. 41 The brain is unique in that it has no capacity to store energy, and 

therefore, it needs a consistent supply of glucose and oxygen-rich blood. Experimental studies in 

rats have shown that following mTBI, there is a decrease in CBF by as much as 71% in injured 

areas.42 Importantly, this study also demonstrated a relationship between the degree of CBF 

reduction and the degree of delayed axotomy, a form of axonal injury that is another critical aspect 

of mTBI pathophysiology. 

Traumatic axonal injury (TAI) is a description to delineate a developing area of research into 

secondary axonal injuries following TBI.43 The exact mechanism of axonal injury has long been 

debated. The cause of morphological changes to axons observed after injury, such as neurofilament 

disorganization and blebbing, a bulging of the cell membrane, was a point of contention.44Kandel 

 It was debated whether this was the product of primary physical damage or a result of 

secondary injury. Classical feline studies demonstrate that injections of horseradish peroxidase do 

not indicate a change in axonal membrane permeability to macromolecules following mTBI.44 The 

same study demonstrated that mTBI caused neurofilament misalignment and axonal blebbing or 

swelling. This was hypothesized to be due to the influx of calcium and activation of Ca2+-induced 

proteolytic pathways, as opposed to the physical forces of the injury. Indeed, unbound intracellular 

calcium above a threshold concentration can activate phospholipases, proteases, and nucleases.45 

More recent work in vitro has demonstrated that, in fact, intracellular calcium does play an 

important role in axonal injury in mTBI. Interestingly, however, it seems the release of intracellular 

stores of calcium is the initial perturbation leading to calcium-dependent cytoskeletal damage and 

morphologic changes.46  

Experimental animal models of mTBI have demonstrated that the hours and days after 

mTBI hold incredible challenges to neuronal cell homeostasis. Upon impact, potassium flux, 

excitatory neurotransmitter release, and subsequent depolarization disturb ionic equilibrium. In an 

  7 

attempt to restore ionic equilibrium, ATP-dependent ion pumps consume energy, moving ions out 

of the cell This leads to an increased demand for glucose metabolism in the mitochondria. 

Mitochondria are attempting to both sequester cytotoxic calcium and generate ATP, and cannot 

meet the cell’s energy needs. These challenges at the cellular level, seem to last 7-10 days after injury 

in animal models.33 They represent a molecular and cellular explanation for the window of cerebral 

vulnerability (WoCV), a description recently coined in the human literature.47 The major peril during 

this window is a risk of reinjury due to the fragile metabolic state of the cell and accumulating 

cytoskeletal damage. In the weeks and months following mTBI, further cellular damage occurs 

through inflammation, intensified due to damage sustained to the blood-brain barrier. 

A major cause of secondary injury post-mTBI is damage accumulated due to the activity of 

the innate immune system. The brain has its own immune cells, microglia, which have been 

implicated in initiating deleterious immune responses following mTBI.48 Following mTBI injury, 

microglia are attracted to sites of neuronal cell damage by chemokines released by astrocytes. 

Notably, this process is a normal physiologic process to quarantine damaged tissue. There appears to 

be differential activation of microglia, with some aiding in tissue repair and others contributing to its 

damage.49 Nonetheless, there is evidence of astrogliosis and subsequent microglia recruitment after 

TBI, as demonstrated by glial fibrillary acidic protein staining (GFAP) and staining with monoclonal 

antibodies ED1 and OX42.50 These data should be interpreted cautiously, as these monoclonal 

antibodies do not stain specifically for microglia and will also demonstrate other 

monocyte/macrophage-like cells. Indeed, other cell types, indistinguishable from microglia, are 

present in the injured tissue following mTBI. This is proposed to result from an inflammatory-

mediated breakdown of the BBB following mTBI. Leukocyte invasion of brain tissue appears to be 

an essential event in perpetuating adverse effects of the post-TBI immune response.51 

  8 

The BBB is a highly selective physical barrier between the bloodstream and neural tissue. It 

is maintained to prevent the entrance of immune cells and neurotoxic substances. Additionally, it 

allows for the transfer of toxic byproducts of neuronal cell function to re-enter the bloodstream for 

excretion.40 In mTBI, byproducts of the inflammatory response, including reactive oxygen species 

(ROS) and Interleukin-1β (IL-1β), result in upregulation in matrix metalloproteinases (MMPs).51 

MMPs are endopeptidases and are essential for extracellular matrix turnover. It has been 

demonstrated in rats that MMP-9 expression, as measured by gelatin zymography and reverse 

transcription polymerase chain reaction, is upregulated by a ROS-dependent mechanism.52 In 

addition, it was demonstrated in a rat model of cerebral ischemia that MMPs alter the structure of 

occludin and claudin, tight junction proteins essential for BBB integrity.53 Taken together, it appears 

that inflammation, as a result of neuronal cell damage, has the potential to cause damage to the BBB 

through disruption of its microstructure. This breakdown of the BBB allows leukocyte infiltration of 

brain tissue and further perpetuates inflammation.  Interestingly, it is known that MMPs activate 

pro-interleukins or cleave pro-interleukins, potentially igniting a vicious cycle of inflammation and 

BBB damage.54 

Post-concussion syndrome or persistent post-concussive symptoms are interchangeable 

classifications for individuals with a mTBI who have symptoms for a period longer than what is 

considered a normal time for recovery. Arbitrarily, this is usually thought to be two to three 

symptoms that are worse than before the mTBI injury, and lasting greater than 3 months post-

injury. Commonly, patients experience migraine-like symptoms, disruptions in learning and memory, 

and slower reaction time. It is theorized that headache following mTBI reflects altered cerebral 

excitability due to the ionic flux discussed previously.55 A human study in severe TBI has 

demonstrated depressed cortical electrical activity, as measured by electrocorticogram, a pattern 

similar to that observed in migraine.56 A study in mice demonstrated a single mTBI resulted in 

  9 

learning impairments, and mice subjected to multiple impacts exhibited cognitive, learning, and 

behavioral deficits.57 Additionally, MRI studies in rats provide strong evidence that these changes are 

due to altered neurotransmission. For example, administration of antagonists of the NMDA and α-

amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPA) were shown to alter 

electrical and hemodynamic activity in rats significantly.58 This may also be responsible for changes 

in reaction time identified in humans. It has been shown using a novel device to measure reaction 

time in humans; reaction time is significantly slower in individuals suffering from mTBI.59 

Other major post-concussive symptoms are behavioral disturbances such as sleep 

disturbance, anxiety, and depression. In a population-based sample of mTBI patients, 65% 

experienced sleep difficulties in the 2-weeks following injury, and 41% had clinically significant sleep 

disturbances one year after injury. 60 It is hypothesized that sleep disturbances have a synergistic 

effect on the negative inflammatory response, potentially increasing the time to recovery in mTBI 

patients. 61,62  

The main observations in animal models of TBI are shorter wake periods and increased 

sleep bouts. 63 The evidence for a synergistic effect of sleep and damaging inflammation is 

extrapolated from work done outside of the TBI literature; nonetheless, it is a major complaint 

following mTBI and warrants further investigation. Finally, changes in neurotransmission in the 

basolateral amygdala and hippocampus, important limbic structures in fear conditioning, are 

hypothesized to be responsible for an increased incidence of anxiety following mTBI. Animal 

studies have demonstrated increased NMDAR expression in these limbic structures and decreased g-

aminobutyric acid-mediated inhibition. These changes were associated with increased fear 

conditioning and anxiety-like behaviors in multiple fear conditioning paradigms 64,65. Management of 

the underlying causes of these symptoms are important targets of ongoing research to improve the 

clinical management of mTBI. 

 10 

For most, the symptoms of mTBI fade after approximately 6-months. More recently, there 

has been growing concern that the resolution of symptoms is not the end of this story. In fact, it has 

been demonstrated in humans that inflammation following mTBI can persist for years after the 

initial injury.66 It is hypothesized that J2-prostaglandins (PGJ2) might mediate the transition from an 

acute and beneficial inflammatory response to a chronic and damaging one. After injury, PGJ2 is 

released by astrocytes and microglia, exhibiting both anti- and pro-inflammatory effects. A review of 

the experimental literature depicts PGJ2-mediated disruption of mitochondria and the ubiquitin-

proteasome pathway as a potential mechanism for this switch.67 

More recent experimental work has advanced the PGJ2 hypothesis. An in vivo study 

demonstrated that a particularly reactive 15-deoxy-PGJ2 caused activation of the integrated stress 

response (ISR) via phosphorylation of eukaryotic translation initiation factor 2 alpha.68 The ISR is an 

adaptive cellular pathway activated in precarious situations to restore cellular equilibrium.69 However, 

long-term activation can lead to cellular damage and even programmed cell death.69 Interestingly, 

animal models of mTBI have demonstrated increased cyclooxygenase (COX) concentration, the 

rate-limiting enzyme in prostaglandin production, in injured neurons.70,71 Chronic inflammation and 

resultant BBB breakdown are thought of as a potential link to what is arguably the most detrimental 

long-term effect of repetitive mTBI, chronic traumatic encephalopathy (CTE).72 

The molecular and cellular events following mTBI are complex, and any treatment must 

consider the multiple perturbations to neuronal homeostasis that occur following mTBI. This 

challenge is particularly evident given the numerous clinical trials (over 30) that have failed to 

produce meaningful patient results.51 Additionally, the cellular and animal literature itself is 

challenging to translate due to a variety of experimental models of mTBI, e.g. fluid percussion, 

weight drop, etc., and no objective criteria to delineate how to produce a ground truth for the 

severity of simulated mTBI.73,74 Therefore, great effort is being placed into defining the in vivo 

 11 

pathophysiology of mTBI in humans fully, using current experimental knowledge as a guide and in 

vivo methods such as neuroimaging before attempting to develop more specific model systems. 

Currently, mTBI patients are diagnosed and monitored clinically with neuropsychological 

tests, but primarily by patient reports of symptoms.75 Close medical observation and intervention are 

only used if more severe pathology is suspected, e.g., cerebral hemorrhage. However, many patients, 

including those with mild injuries, often receive neuroimaging in the form of a routine computed 

tomography (CT) scan of the head. The treatment for a sports-related mTBI consists of rest and a 

gradual return to mental and physical activity. When symptoms disappear, patients are considered 

“recovered” from a mTBI, which may or may not coincide with neuropsychological measures 

returning to baseline values. However, based on the combined human and animal literature 

presented, recovery cannot be definitively determined by symptom resolution alone. There is a great 

need for objective criteria to monitor true biological recovery from mTBI.  

A clinically useful objective marker for diagnosis of mTBI will reflect cellular damage to 

neurons or related aspects of mTBI pathophysiology. The quest for a measurable prognostic blood 

biomarker of mTBI is an ongoing search. A recent review points towards numerous blood 

biomarkers that have shown promise.4 For example, neurofilament light chain (NfL), a marker of 

neurofilament damage due to axonal injury, and glial fibrillary acidic protein (GFAP), a marker of 

astrogliosis, both were found to be elevated in patients suffering mTBI compared to controls.13,76 Of 

particular interest, given rising concerns about CTE following mTBI, are biomarkers that would link 

mTBI pathophysiology to the tauopathy’s characteristic of CTE. In a recent animal model of 

tauopathy, it was demonstrated that NfL is not only elevated in CSF and plasma, but it also appears 

to be a good marker of response to treatment with β-site amyloid precursor protein cleaving 

enzyme-inhibitor.77 

 12 

A recent study developed a new highly sensitive immunoassay for NfL and examined its 

efficacy as a biomarker for mTBI in a sample of professional hockey players. They demonstrated 

that NfL was correlated with a standard of care neuropsychological test and was effective in 

predicting return to play time.78 NfL outperformed other biomarkers, such as S100 calcium binding 

protein B, neuron-specific enolase, and plasma tau, in predicting return to play time, a newer criteria 

for severity grading in sports-related mTBI. Several other studies in athletic populations have 

demonstrated increases in plasma NfL following even sub-concussive hits.79–81 Ideally, in future 

studies, serial and longitudinal sampling would be conducted to determine if there is any circadian 

fluctuation and begin to gain an understanding of how plasma levels of NfL change after injury. The 

culmination of years of blood biomarker work more recently resulted in the FDA approval of 

GFAP and UCH-L1 for plasma detection of intracranial injury following mTBI.76,82 However, there 

still remains a need for a biomarker of true biological recovery, and an understanding of the 

longitudinal changes in these markers in recovery. For these reasons, it is important that a paradigm 

shift take place. mTBI should not be thought of as an accident with a swift resolution, but as a 

potential trigger for a long-standing cascade of pathophysiology that ultimately can lead to 

neurodegeneration. Some even feel that ‘mild’ is a designation that should be entirely done away 

with, and that we should focus efforts on outcome measurement instead of imprecise classification 

systems.83  

Presently, mTBI is a clinical diagnosis guided by a thorough history and physical exam 

conducted by a physician, with imaging and clinical assessment scores as supplements. A traumatic 

brain injury is classified as mild based on the widely accepted American Congress of Rehabilitation 

Medicine diagnostic criteria for mTBI, which was recently updated in 2023.84 To be considered a 

mild injury, the patient must have a plausible mechanism of injury likely to cause trauma, and meet 

one to two of the following criteria: 1) have one or more signs attributable to the injury; 2) at least 

 13 

two symptoms and at least one clinical laboratory finding attributable to the injury; and 3) CT or 

MRI intracranial abnormalities attributable to the injury. Patient diagnosis is elevated to a higher 

severity if they meet the following criteria: 1) loss of consciousness greater than 30 minutes; 2) 

Glasgow Coma Scale less than 13, 30 minutes after the injury; and 3) post-traumatic amnesia greater 

than 24 hours. Signs of injury are considered observable acute changes in the physiological 

functioning of the brain attributable to the injury, including loss of consciousness, altered mental 

status, amnesia for events, and other neurologic signs such as seizure or motor incoordination. 

Symptoms, on the other hand, are divided explicitly into somatic, such as dizziness or headache, as 

well as neuropsychiatric, such as anxiety or difficulty focusing. Ultimately, in the office, the 

resolution of clinical symptoms and an individual's ability to return to work, play, or learning 

environments drive the clinical resolution of the injury.   

Although an injury may be classified as mild, clearly, this is an imprecise description.83 In the 

future, the clinical and basic research communities will need to focus on understanding what drives 

differences in outcomes among individuals who are considered to have a mild injury. Recent work 

from the TRACK-TBI study demonstrated that up to 54% of patients following one year of injury 

were experiencing functional limitations, and many reported greater than one or more symptoms 

that were worse before their injury. Patients experiencing this type of outcome are far from mildly 

injured.   

DIFFUSION MAGNETIC RESONANCE IMAGING 

The following section will give a foundational overview of how the DWI signal is created. 

This information will be given using a conceptual framework aimed at a neuroscientific and clinical 

audience, with few underlying mathematical equations included. An understanding of fundamental 

magnetic resonance physics and spin magnetic resonance is assumed. For a more mathematically 

based explanation, I suggest the first four chapters of the text Diffusion MRI or the third chapter of 

 14 

 
 
Introduction to Diffusion Tensor Imaging for a more abridged explanation.85,86 Notably, the 

underlying physics principles are complex, and some simplifications has been utilized to aid in 

understanding. In addition, the data collection process and different vendor equipment are nuanced; 

a clinician or basic scientist should not undertake an MRI experiment without the direct 

involvement of an MRI physicist to facilitate data quality and harmonization. George Fueschel 

coined the phrase, “Garbage in, Garbage out”, which most certainly applies here.  

MRI allows for the observation of signals from nuclei; when discussing DWI, the nuclei in 

question are the proton (1H) of water molecules in the human body. The experiment begins by 

putting energy into the subject tissue (in this document, the brain) and listening for the signal 

emitted after the excitation, referred to as an echo. The signal itself is caused by a current induced by 

magnetic flux due to the precession of spins around a static magnetic field and detected as a voltage 

by a receiver coil. The information in the signal has a frequency, intensity, and phase produced by 

the magnetic field oscillations. This data is collected at a particular time after the excitation.  The 

composite data, called k space is then transformed to become an image via Fourier Transform.87 

The opening of the MRI, or bore of the magnet, has a continuously active magnetic field 

oriented through the bore or along its z-axis, referred to as B0. The x-axis is in the left-right 

direction, and the y-axis is in the up-down direction. When a subject’s brain enters the magnet's 

bore, all of the spins of the protons hydrogen nuclei of water molecules are oriented along this B0. 

The signal of interest during experiments is not generated with by B0 magnetic field, but rather as a 

result of secondary magnetic field gradients applied in transverse planes to B0.88 By linearly 

modulating the B0 field strength along any of these axes or a combination of them, the spins can be 

tipped out of alignment with B0. Using combinations of gradients with different orientations and 

strengths, we can observe signals where certain relaxation times, for example, T1 or longitudinal 

 15 

relaxation time, are emphasized. Relaxation times are specific to different tissues, so depending on 

the scientific question, there is great variability in the actual makeup of an MRI experiment.  

Using the frequency, intensity, and phase of the emitted signal, scientists and clinicians can 

observe the underlying anatomy of the brain, its function, and its physical connections. Collectively, 

the program of changing magnetic gradients that produce the signal are referred to as “pulse 

sequences,” of which DWI (diffusion weighted imaging) is a specific type.89 These pulse sequences 

have various parameters that an MRI Physicist can manipulate to produce a particular tissue 

contrast. Pulse sequences can be made sensitive to specific tissue-dependent characteristics in the 

case of T1 or T2 weighted images. Alternatively, they can be used to measure signals sensitive to an 

underlying phenomenon, for example, diffusion or blood oxygen-dependent signal of functional 

MRI. What is unique about DWI is that it is not necessarily the underlying tissue characteristics we 

are measuring directly but rather the alterations in the random motion of water in the human brain 

caused by the organized tissue structure of white and grey matter. 90 

The diffusion of water molecules we are interested in measuring is averaged within the 

confines of the approximately 2mm isotropic “voxel”, a three-dimensional volume from which we 

are measuring data, and is also referred to as intravoxel incoherent motion or, more commonly, 

Brownian motion.91 This is distinctly different than bulk flow of a fluid, for example, the movement 

of blood within the internal carotid artery. Typically, the motion of a single water molecule during 

the time of a single diffusion-weighted pulse sequence is around 1-20 µm, significantly smaller than 

the voxel from which we measure, and, therefore, it is essential to reiterate that we are observing the 

averaged motion of large quantities of water molecules.90 We use gradients that make the signal 

sensitive to the diffusion of water molecules. In DWI, multiple gradients with varying orientations 

are used to acquire information about the underlying diffusion directionality of water molecules. 

 16 

All molecules, at temperatures above absolute zero, possess thermal energy that results in 

motion, commonly called molecular diffusion.91 Work by Albert Einstein (1.1) proposed that the 

conditional probability distribution of a group of particles after a diffusion time (t) is proportional to 

the diffusion coefficient (D) where (x2) is the mean squared displacement in one dimension.92 

𝑥! = 2𝐷𝑡 

(1.1) 

The diffusion coefficient (D) can be thought of as an innate property of the tissue being measured 

that depends on the size of the diffusing molecules (in this dissertation, water or H2O), the 

temperature, and the microstructural environment in which it is being measured.  

Figure 1.1 – A vector magnetization diagram or pulse sequence diagram representative of the spin-
echo sequence described by Stejskal & Tanner in 1965. Diffusion Gradients (DG) are applied to 
“label” protons of water molecules. Those molecules that move from their original position will fall 
out of phase (red line), resulting in signal loss. Courtesy of Allen D. Elster, MRIquestions.com 

A typical DWI pulse sequence shown in Figure 1.1 builds on a spin-echo sequence, which 

gives a T2-weighted contrast.93 Following the 90o radiofrequency pulse, the first diffusion-sensitizing 

gradient (DG), is applied, causing different protons to experience different magnetic field strengths 

and resonate at different frequencies depending on their location. This can be thought of as 

effectively labeling all protons in a single brain slice. Then, after the second 180o radiofrequency 

 17 

 
 
 
 
(RF) pulse, a second diffusion-sensitizing gradient, is applied.93 The approach was adapted from a 

body of work on characterizing diffusion with nuclear magnetic resonance and is called Stejskal-

Tanner diffusion encoding.94 Figure 1.1 represents a single repetition of the acquisition of data in k-

space. This can be thought of as acquiring data in a single three-dimensional slice of brain tissue and 

is repeated many times to acquire DWI data in slices that cover the entire brain. 

Protons that have not moved or diffused since the initial dephasing gradient will regain the 

same phase and emit a signal as indicated by the green line in Figure 1.1. However, protons that 

have diffused or moved from their original position will fall out of phase, resulting in overall signal 

loss when combining the effect of all protons in a voxel. Recalling (Equation 1.1), water molecules 

allowed to diffuse in an unrestricted environment for a certain time will diffuse equally in all 

directions; this is termed “isotropic” diffusion.91 The greater the diffusion distance, the greater the 

signal loss. The tissue type in the brain with the greatest diffusion distances is the CSF, and therefore 

experiences the greatest signal loss and appears dark on DWI images.  

Most, if not all, molecules in the brain have moved away from their original position, 

resulting in a spread of phases and attenuation of the signal. However, in the highly organized WM, 

diffusion distances are far greater along the “grain” of the WM, as opposed to across bundles of 

axons.95 This is due to the tightly packed nature of the WM pathways and the negligible diffusion 

across myelin sheaths or axon cell membranes. Therefore, the diffusion is said to be “anisotropic” in 

WM or not the same in all directions.91 This process of applying gradients is repeated multiple times 

along many directions to capture the complex fiber architecture of the brain's WM pathways.93 Each 

additional gradient direction helps to detect the direction of diffusion and improve the accuracy of 

diffusivity quantification, which infers changes in the microanatomy and organization of 

connections of WM pathways.  

 18 

Another important concept when discussing DWI is the concept of a “b-value” and how it 

relates to the diffusion-weighted signal. Mathematically, the two are related with (Equation 1.2):   

𝑆 = 𝑆"𝑒#$% 

(1.2) 

The signal without diffusion weighting (S0) is related to the signal with diffusion weighting 

(S) by the term (e-bD) in which (D) represents the diffusion coefficient and (b) represents the b-factor 

or b-value.91 At this point, a clinician will appreciate noting that the equation term (e-bD) when no 

diffusion gradient is utilized, would leave only T2-weighting (e-TE/T2) of the signal, as mentioned 

previously. DWI sequences have a long echo time (TE), and, for this reason, are heavily T2-

weighted. Therefore, the first few images in a DWI dataset without the diffusion-weighted gradients 

applied, referred to as b=0 images (pronounced b-zero), look very similar to a typical T2-weighted 

image where CSF appears bright.96  

Quite simply, the b-value represents how sensitive the pulse sequence is to diffusion. The b-

value is proportional to the square of the gradient strength; stronger gradients (higher b-values) are 

sensitive to smaller diffusion distances.93 Equation (1.3) represents the b-value for a rectangular 

gradient where (G) is the strength of the diffusion gradient, (d) is the length of the gradient, and (D) 

is the length of time between the onset of the first gradient and the onset of the second gradient.   

𝑏 = (𝛾𝐺∆)!(∆ −

𝛿
3

) 

(1.3) 

The b-value has units of seconds per millimeter squared. Gradient strength, length of the 

gradient, and length of time between gradients are parameters that an MRI physicist can manipulate 

to alter the diffusion weighting of a specific pulse sequence.93 This is at the expense of signal-to-

noise ratio (SNR), with higher b-values having lower SNR, as evident by the relationship between 

equations 1.2 and 1.3. 

 19 

 
 
 
 
In the present dissertation, data with two separate b-values, or shells, will be utilized. This is 

important to mention because these two shells were collected separately at the time of data 

collection. From a practical standpoint, when a b-value is chosen for a study, the diffusion time is 

typically not set purposefully. Instead, it reflects the shortest duration of gradient application that is 

possible to achieve the desired diffusion weighting. This means diffusion times may vary between b-

values, and as a result, two separate acquisitions at different b-values may have slightly different 

repetition times (TR). Ultimately, this results in slight differences in signal intensity, which can be 

problematic when processing concatenated data with different b-values. This is only an issue when 

using modern modeling techniques that use both shells instead of one to estimate diffusion metrics. 

We address these slight differences in signal intensity between shells using a data harmonization 

approach described in Chapter 3 of this dissertation.      

Finally, we would like to briefly touch on the strengths and limitations of the specific pulse 

sequence used for the TRACK-TBI data acquisition. Understanding the specific pulse sequence 

used to collect DWI data is crucial for clinicians designing studies as it will dictate the type and 

severity of artifacts and noise present in the data. TRACK-TBI utilized a cutting-edge acquisition at 

the time of the study's inception, known generally as a multislice single-shot spin-echo echo planar 

pulse sequence.15 The data were collected on a General Electric (GE) MRI scanner, for which the 

proprietary name of the sequence is single-shot fast spin echo echo planar imaging (EPI). Early in 

the clinical use of DWI, data was collected using single-shot echo-planar imaging (EPI) methods, 

which are faster due to the use of EPI echo train.97 However, data acquisition with EPI echo train is 

prone to susceptibility field distortions, especially in regions close to an air-tissue interface.98 A 

strength of the diffusion EPI methods was their resilience to patient motion. This lays the 

foundation for the rigorous preprocessing necessary before analysis of DWI data outlined in 

Chapters 3 and 4.  

 20 

DIFFUSION TENSOR IMAGING IN MILD TRAUMATIC BRAIN INJURY 

DTI is a modeling technique developed at the NIH largely due to the work of Dr. Peter 

Basser and colleagues.100,101 Interestingly, a pork loin was initially utilized to develop this technique, 

and it was not initially conceived for studying the WM.102,103 This subtlety, although seemingly trivial, 

is actually quite important, as a major weakness of this technique when applied to human WM is an 

inability to model the complex architecture of certain WM regions, which is noted in the original 

manuscripts published in 1994. Despite the excitement at the time in the neuroscience community, 

this limitation makes the technique ill-suited for studying the brain's WM comprehensively. 

Additionally, fiber tracking based on the DTI model is problematic, as any voxels containing signal 

contributions from more than one distinct fiber population could result in inaccurate 

connections.20,104 Nonetheless, due to its ease of use and sensitivity to pathologic changes in many 

disease states, the neuroimaging community widely adopted it with success, albeit the interpretation 

of findings was questionable.105 Some have argued its simplification of the complex information 

collected with DWI is, in fact, a strength as opposed to a weakness.20 I propose it is an emerging 

scientific controversy whether or not DTI should continue to be used as the sole modelling 

technique in mTBI studies due to well documented limitations of the model.1–4 This section will 

overview the diffusion tensor model, the localization of effects, and patterns of changes observed in 

DTI metrics in the WM of mTBI patients.  

The diffusion tensor model is represented by equation (1.4) and is applied to each voxel of 

diffusion MRI data collected to fit numerous tensors throughout the brain.100 This equation relates 

the signal (S0) from the b0 image without a diffusion gradient applied to the signal (Sj) with a 

diffusion gradient (j) applied.106 The signal with a gradient applied is equal to the signal without a 

diffusion gradient applied multiplied by the exponential value of the known b-value (b), a unit vector 

 21 

 
(xj) representing the direction of the applied gradient, and the unknown value of the diffusion tensor 

D.  

𝑆& = 𝑆"𝑒𝑥𝑝2−𝑏𝑥&

⬚
’𝐷𝑥&3

(1.4) 

The diffusion tensor is a matrix with dimensions equal to the number of gradient directions applied, 

with the lowest possible being six unique directions and, therefore, a three-by-three matrix.100  

The diagonal elements of the tensor along the x, y, and z-directions are in scanner space. 

These are what a diagnostic radiologist may be familiar with seeing when discussing diffusion-

weighted MRI and are used to calculate the Apparent Diffusion Coefficient (ADC). These diagonal 

elements are averaged to create the trace image or diffusion-weighted image commonly used with 

the ADC image in diagnostic imaging of stroke patients.107,108 The ADC map is generated by dividing 

the signal in each voxel of the trace image by the signal in each voxel of the b0-image and taking the 

logarithm of that result. Therefore, the ADC and trace images are inversely related, with restricted 

diffusion, as in the region of an infarct, appearing dark on the ADC image and bright on the Trace 

image.107  

For DTI, more commonly used in research applications, the tensor information in scanner 

space is diagonalized using an eigendecomposition where the direction of principal diffusion, or 

eigenvector (v1), corresponds to the direction of greatest diffusivity, which theoretically should be 

along the “grain” of the WM.95 The shape of the diffusion tensor is described as either isotropic, 

with diffusivity equal in all directions, and therefore, all eigenvalues are similar, or anisotropic 

diffusivity, i.e., not equal in all directions.91 The diffusion tensor is anisotropic in WM, owing to the 

coherent organization of tightly packed myelinated axons. Likewise, in the less organized grey 

matter, the diffusion tensor is isotropic; however, it is obviously less so when compared to CSF.    

 22 

 
 
 
 
The eigenvalues can then be used to calculate quantitative diffusion maps or diffusion tensor 

metrics/scalars, such as the summary measures fractional anisotropy (FA) and mean diffusivity 

(MD), two commonly published metrics. FA is the variance across the eigenvalues with a value of 0 

to 1, with one representing completely anisotropic diffusion and 0 representing isotropic diffusion.109 

MD is the average of the eigenvalues, with WM generally having lower MD when compared to grey 

matter and CSF.105 Less commonly reported DTI metrics include Axial diffusivity (AD), which is the 

diffusivity along the principal direction of diffusion and is equal to the first eigenvalue (l1). Finally, 

radial diffusivity (RD) is the average of the second (l2) and third eigenvalues (l3), perpendicular to 

the first, representing the diffusivity across the grain of WM.  

Commonly in mTBI, DTI metrics are interpreted as reflecting changes in WM 

microstructure, with the most common pattern observed in the acute period following injury being 

decreased FA and increased MD.3,5,17 This change after mTBI is interpreted as being related to 

diffuse axonal injury, associated edema, and increases in cellularity during the ensuing inflammatory 

process. These claims have been confirmed in ROI analyses with coinciding histopathology in 

animal studies examining colocalized voxels with DTI metrics and histologic staining.110,111  

The TRACK-TBI Study, the main source of data for this dissertation, collected various MRI 

pulse sequences in over 300 mTBI patients recruited from level 1 trauma centers across the U.S., 

specifically collecting DWI data at 2-weeks and 6-months post-injury.112 The study was originally 

designed to create a set of common data elements related to outcome measurement for traumatic 

brain injury in a population-based sample recruited across the US. Therefore, a comprehensive 

battery of 21 unique outcome measures was collected, including commonly used measures like the 

GOSE and the RPQ. The study also collected various combinations of single-shell, multi-shell/high 

angular resolution diffusion imaging (or HARDI data) at various b-values.113 

 23 

An early study on DWI data collected during the TRACK-TBI pilot study built on previous 

strong evidence and demonstrated that DWI is sensitive to WM changes following mTBI, even in a 

population-based sample.15 Compared to control subjects this voxel-wise tract-based spatial statistics 

(TBSS) analysis of DTI metrics revealed regions of decreased FA in the internal and external 

capsules, genu of the corpus callosum, uncinate fasciculi, and anterior corona radiata bilaterally. The 

analysis identified no other differences between the control and mTBI subjects. The sample of 

mTBI patients was subsequently divided into CT/MRI negative (n=44) and positive (n = 32), and 

comparisons of DTI metrics in these two groups did not differ significantly.15 DTI metrics also did 

not differ when comparing the upper and lower halves of the control group dichotomized by years 

of education. DTI metrics have been shown to covary with factors like age and years of education in 

other samples.120, 178 

In the same study, a post-hoc ROI analysis was conducted using the intersection of regions 

in the Johns Hopkins University WM atlas and clusters of significance from the analysis of the FA 

skeleton generated during the TBSS analysis.114 If a voxel fell 2.2 standard deviations above or below 

the mean of the control group ROIs it was considered abnormal.15 Interestingly, a subgroup of 

mTBI patients with a negative CT/MRI on the day of injury had a trend-level difference between 

the proportion of abnormal ROIs (25%) compared to controls (10%). The proportion of abnormal 

ROIs in the CT/MRI positive group (43.8%) did not significantly differ from that of the CT/MRI 

negative group (25%). This nicely highlights that even amongst a group of patients categorized as 

mild, there is heterogeneity in the proportion of injured tissue, which, importantly, is not detectable 

by the current standard-of-care imaging techniques. Also, in this early study, a point not made by the 

author was a lack of observed difference in the GOSE scores of the CT/MRI-negative group and 

CT/MRI-positive group at 6-months post-injury, with 56% and 59% reporting a GOSE <8 in the 

 24 

 
CT/MRI-negative vs positive groups respectively. This suggests there is an additional mechanism 

driving recovery beyond resolution of what matter damage localized with DTI. 

Subsequently, another DTI analysis was conducted on the whole TRACK-TBI sample using 

single shell DWI data at b = 1300 mm/s2, with 367 mTBI patients included in the final analysis.115 

The study observed significantly decreased FA at the 2-week point in the external capsule, superior 

fronto-occipital fasciculus, and genu of the corpus callosum when comparing mTBI patients to 

(n=148) friend controls. Conversely, the mTBI group had higher AD, MD, and RD in the cerebral 

hemispheres at the 2-week time point compared to controls. Post-hoc tract-wise analysis utilizing 

the JHU atlas revealed large effect sizes, with the largest approaching and exceeding 1 for AD in the 

body of the corpus callosum, the superior longitudinal fasciculus, the external capsule, the anterior 

limb of the internal capsule and the superior corona radiata.  

Longitudinal change was also examined in this second TRACK-TBI study of DWI data, 

although it needs to be clarified how this was done. This analysis noted no significant changes in FA 

or RD between the 2-week and 6-month time points.115 The stability in metrics from the subacute to 

the chronic stage is a novel and interesting finding that is difficult to explain. A possible explanation 

is the significant proportion of patients with prolonged recovery in this sample, suggested by work 

examining outcome variables from the sample.8 In a comprehensive analysis of outcomes collected 

in the TRACK-TBI Study, 53% of the mTBI sample reported functional impairment at 12 months 

post-injury. CT/MRI negative mTBI patients at the 12-month timepoint had over two times 

increased risk of reporting headache (RR: 2.87), irritability/anger (2.39), frustration (2.67), 

forgetfulness (2.76), and slowed cognition (2.42) compared to controls.7 Additionally, the CT/MRI 

negative group had a slightly increased risk of reporting functional impairment at work 6 months 

following injury compared to controls and reporting mental/behavioral changes that affect their 

close relationships at 12 months post-injury. It is important to note here that these are symptoms 

 25 

 
commonly reported in the general public, and can be due to factors other than TBI as noted in a 

response to the original article.116 However, the instrument used (GOSE) was specifically modified 

for this study to specifically ask about symptoms related to the mTBI, and the RPQ asks patients to 

rate symptoms based on pre-injury levels.83 Nonetheless, data collection at a level 1 trauma center 

may have contributed to this sample being more severe on the imprecise spectrum categorized as 

mild. This would seem to suggest that damaged white matter has only a marginal capacity to heal 

following mTBI and does not return to a normal defined by the control subject mean.  

Studies utilizing DTI in the extant literature observe heterogenous patterns when compared 

to the TRACK-TBI study result, and some studies that report similar patterns of change in DTI 

metrics have failed to correlate these findings to outcome measures.3,17 The design of DWI studies 

can be challenging and interpretation of DTI results must be done cautiously. Many changes to the 

environment in which water molecules are diffusing can alter DTI metrics.105,117  For example, 

observations of higher FA in the acute/subacute period following mTBI are often made in pediatric 

populations.17 This could be due to heterogeneity in scan timing following injury, analysis methods, 

and DWI acquisition differences.118,119 Alternatively, there could be a heterogenous response in WM 

within mTBI patients that we have failed to quantify adequately. Beyond that, differential effects as a 

result of heterogeneity in force distribution from the injury and varying degrees of the post-injury 

pathophysiologic cascade in individuals may also be at play. This is all complicated by the field 

actively moving from simpler DWI acquisitions and DTI analysis to the use of HARDI techniques 

and novel higher-order modeling techniques developed to take advantage of this DWI acquisition, 

which is not necessarily comparable to classical diffusion studies in mTBI.104,120 

The diffusion tensor model to date has been useful, and clearly, DWI data modeled with 

DTI can detect WM changes following mTBI that conventional imaging methods cannot. There are 

limitations, however, to the interpretability of the results inherent to the way in which the DTI 

 26 

  
scalars are calculated. For example, reduced FA reported in the TRACK-TBI Study could arise from 

several mathematical scenarios. FA could be changing due to a reduction in diffusivity along the 

principal diffusion direction aligned with the grain of the WM or, equivalently, a reduction in the 

first eigenvalue.121 Similarly, an increase in diffusivity across the grain of the WM (increased RD) 

could decrease FA, and finally, a combination of these two could occur.122 These alterations in the 

DWI signal represented by the DTI scalars calculated in this fashion could be caused by many 

combinations of underlying changes to axons and their environment. Early animal studies utilizing 

DTI demonstrated increased diffusivity across the grain of WM (increased RD) in shiverer rats who 

have mutated myelin basic protein, resulting in malformed or absent myelin sheaths.123 Later work 

demonstrated that axonal degeneration preceding demyelination in mice fed cuprizone was 

associated with a decrease in diffusivity along the grain of WM (decreased FA) with no change in 

diffusivity perpendicularly.124 Nonetheless, in mTBI, a combination of multiple pathophysiologic 

mechanisms with different time courses makes interpretability a challenge for DTI metrics.  

In our own lab, past work has also demonstrated that other quantitative MRI sequences may 

be sensitive to the effects of mTBI on the brain. In a sample of sport-related concussion/mTBI 

athletes at MSU, a significant drop in default mode network connectivity was observed in the 

subacute recovery phase following concussion, which was correlated with various composite scores 

of the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) 

neuropsychological test battery.125 Therefore, combinations of structural and functional quantitative 

MRI techniques, so-called multimodal techniques, will likely aid in a comprehensive understanding 

of brain changes following mTBI.3 After decades of research using DTI, questions remain about the 

time course of observable changes in DTI metrics and, importantly, how they relate to the 

underlying pathophysiology and WM microstructure. This has largely led to denouncement of 

interpreting DTI changes as relating to changes in “WM integrity” as the measure does not offer 

 27 

specific information to extract such conclusions.20,104,105,121 Therefore, fundamental questions about 

the localization and extent of damage from mild injury and subsequent development of downstream 

sequelae remain unclear in humans. The largest outstanding question is how locations and extent of 

damage in the brain detectable with DWI relate to alterations in cognitive functioning and 

symptoms. 

 28 

 
 
CHAPTER 2: A RAPID REVIEW OF MILD TRAUMATIC BRAIN INJURY STUDIES 
UTILIZING CONTEMPORARY TECHNIQUES FOR DIFFUSION IMAGING ANALYSIS 

Joshua H. Baker, Andrew Bender, Sarah E. Tilden, Norman Scheel, David C. Zhu1 

1Chapter 2 contains an invited rapid review manuscript that is currently under final review before 

submission to the Journal of Neurotrauma. Its use within the current dissertation was approved by the 

doctoral committee. Joshua H. Baker conceived the original idea, conducted the search for 

manuscripts, reviewed the manuscripts, and drafted this manuscript independently. The manuscript 

has been edited to fit the format of this dissertation. 

INTRODUCTION 

Diffusion-weighted magnetic resonance imaging (DWI) has been used extensively to study 

the brain following mild traumatic brain injury (mTBI). However, several recent comprehensive 

reviews, which have looked at over 100 studies collectively, concluded that a consistent pattern of 

change measurable with this pulse sequence/modality was not observed.3,5,17 This lack of consensus 

may be due in part to individual differences in injury mechanism, components of resilience to injury, 

and even heterogenous phenotypes of WM recovery following injury. However, these inconsistent 

findings may also be partly due to the widespread use of the diffusion tensor imaging (DTI) model.  

DTI is the most common method for modeling the DWI signal in neuroimaging studies of 

mTBI. There is no debate about DTI’s sensitivity to microstructural changes in the WM following 

mTBI, even in patients with negative findings on conventional neuroimaging. However, this 

technique has several limitations, and newer complementary higher-order signal representation and 

modeling methods such as diffusion kurtosis imaging (DKI), neurite orientation dispersion and 

density imaging (NODDI), and Fixel-Based Analysis (FBA) may provide greater specificity in 

studies collecting DWI data in mTBI patients. The present review will introduce the foundational 

 29 

 
 
 
 
principles of these modeling techniques, summarize the metrics estimated with them, and review the 

current findings in studies that utilized each technique in mTBI patients.     

A recent comprehensive review of DTI studies in mTBI concluded the most common 

observed findings are decreased fractional anisotropy (FA) and increased mean diffusivity (MD) in 

WM tracts such as the corpus callosum, internal capsule, and corona radiata.1 This injury pattern is 

thought to reflect the secondary injury cascade related to diffuse axonal injury, inflammatory cellular 

reactivity, and vasogenic and cytotoxic edema following the primary insult. However, there are other 

studies that report higher FA with lower MD and/or radial diffusivity (RD) in the acute/subacute 

period.1 This lack of a reproducible pattern and specificity for secondary injury mechanisms has 

limited DTI’s clinical translation. It is increasingly difficult to ignore the possibility that major 

limitations of the DTI model may be contributing to this ambiguity. The underlying assumption of 

DTI is a Gaussian approximation to the probability distribution governing the random displacement 

of water molecules. Additionally, DTI is a voxel-based technique and has difficulties modelling  

voxels containing anything other than a singular coherent bundle of axons. This limits the sensitivity 

and specificity of DTI in mTBI.126,127  

HIGHER ORDER SIGNAL REPRESENTATION AND BIOPHYSICAL MODELLING 
TECHNIQUES  

Diffusion Kurtosis Imaging 

Diffusion kurtosis imaging (DKI), developed in the early 2000s, characterizes non-gaussian 

diffusion, leading to improved performance in regions with crossing fibers compared to DTI. DKI 

is categorized as a signal model. This is in contrast to biophysical models, which quantify 

microstructural characteristics of the underlying tissues and their change in pathology. The term 

kurtosis refers to the variation in the Gaussian shape of the distribution. The apparent diffusion 

kurtosis is based on the premise that deviations from a Gaussian diffusion probability distribution or 

 30 

 
a sample of data-tailedness in the brain are a measure of underlying tissue microstructural 

organization.128 Mathematically, DKI introduces a quadratic term to the linear relationship between 

the logarithm of the DWI signal and the b-value. Jensen and colleagues reported higher WM 

apparent kurtosis coefficients of 1.41±0.11 compared to grey matter 0.82±0.03, indicating the greater 

organization of structures that will impede free diffusion of water in WM compared to gray matter.1  

DKI data can be characterized by fitting a 3 x 3 tensor matrix with three principal 

eigenvectors and eigenvalues like DTI. The most reported scalar diffusion kurtosis tensor imaging 

(DKTI) metrics are mean kurtosis (MK), which represents the average of the kurtosis along all 

directions; axial kurtosis (AK), the kurtosis along the axial direction of the diffusion tensor; and 

radial kurtosis (RK), the kurtosis along the radial direction of the diffusion ellipsoid (Table 1). DKI 

requires two non-zero b-values and at least 15 diffusion encoding directions.1 The acquisition can 

take 7-10 minutes to use DKTI, or a faster protocol can be acquired in 1-2 minutes to estimate only 

MK.129,129–131 An early study utilizing DKI demonstrated that MK and not MD or FA was sensitive 

to reactive astrogliosis, indicating the method is complementary to DTI.132  

Neurite Orientation Dispersion and Density Imaging (NODDI) 

The neurite orientation dispersion and density imaging (NODDI) compartment model of 

diffusion signal builds on the analytical neurite model developed by Jespersen and colleagues, 

2007.133 This method models the diffusion signal in grey and WM, with the summary term neurite 

referring collectively to axons in WM and dendrites in grey matter. NODDI models the signal from 

three microanatomical locations: the intra- and extracellular compartments and the cerebrospinal 

fluid (CSF); therefore, it is categorized as a multi-compartment model.134 By quantifying the 

contributions from these three compartments and modeling a relationship between them, NODDI 

allows further interpretability of the diffusion signal in addition to DTI and DKI.  

 31 

 
 
The intraaxonal compartment, a unique environment of diffusivity within axons in a voxel, is 

modeled as lines, or cylinders with zero radius. The orientation distribution function is estimated 

using a Watson distribution, which is in contrast to the spherical harmonic series used in the original 

work by Jespersen.133,135 Using a Watson distribution, accurate estimation of scattering dendrites and 

fanning axons (referred to as dispersion), or coherent axon bundles, is possible.  

Notably, compared to previous models, the parallel and perpendicular diffusivities are 

mathematically related. In this framework, increased dispersion and a reduction in intracellular 

anisotropy will also be reflected as a decrease in extracellular anisotropy due to the increased 

perpendicular diffusivity. In the original work by Zhang and colleagues, when the two NODDI-

derived metrics, neurite density index (NDI) and orientation dispersion index (ODI), were plotted 

against stratified FA values, it was shown that different combinations of density and orientation in a 

voxel can result in the same FA value. This additional specificity is likely to aid in untangling the 

inconsistent findings of FA in the acute/subacute recovery period following mTBI.  

NODDI requires a high angular resolution diffusing imaging (HARDI) protocol, with the 

ideal protocol requiring 30 diffusion directions at a low b-value of around 700 mm/s2 and 60 

diffusion directions at a high b-value of around 2000 mm/s2 . The optimized protocol is about 25 

minutes long, while a reduced angular resolution protocol can be acquired in under 10 minutes.134 

The shorter protocol only decreases the certainty in the estimation of the ODI and has little effect 

on the other parameters. The most reported metrics estimated from the NODDI model are referred 

to by many names throughout the literature. They are more commonly the orientation dispersion 

index (ODI), neurite density index (NDI), or isotropic volume fraction (FISO) (Table 1). NDI in 

the original work by Zhang and colleagues was referred to as the volume of the intracellular 

compartment, abbreviated VIC, and may also be seen in various publications as intraneurite fraction 

 32 

 
 
(fi) or intracellular volume fraction (ICVF). FISO, originally, volume of the isotropic compartment 

Viso, is equivalent to free water fraction (FWF/ISO/ffw).  

ODI quantifies how uniform the organization of axons within a voxel is on a scale from zero to one. 

Zero indicates more anisotropic/parallel organization, whereas one represents isotropic/dispersing 

organization. NDI is an estimate of the volume of the intraneurite compartment and, therefore, 

reflects an approximation of axon density. FISO measures the contribution of isotropic diffusion 

present in a voxel, which is normally attributed to CSF. The isotropic compartment serves to adjust 

ODI and NDI, and as a measure of the extent of isotropic diffusion where it does not belong in 

WM. This can be useful in examining the extent to which edema following mTBI may be present in 

WM regions.    

Constrained Spherical Deconvolution and Fixel-Based Analysis 

Constrained spherical deconvolution (CSD) allows for the comparison of multiple metrics 

within a single voxel.18 Importantly, these metrics are specific to distinct fiber populations within a 

single voxel. This new volume element, a single fiber population within a voxel, is referred to as a 

“fixel” (fiber in a voxel). It should be noted that this analysis technique can resolve multiple fixels in 

a voxel.  

CSD extends earlier multi-tensor fitting models by expanding the number of fiber 

populations fit to infinity and neglecting the negative orientations of the fiber orientation 

distribution function.136 Therefore, CSD estimates the fiber orientation distribution function without 

the need for prior knowledge and, importantly, assumes the diffusion characteristics of all fiber 

populations are identical. CSD assumes that variations in WM microstructure have a negligible effect 

on the measured DWI signal and that, instead, partial volume effects are responsible for variations. 

This underlying assumption is what allows CSD to account for crossing fiber architecture. Any 

variations in the diffusion anisotropy are understood to be entirely due to partial volume effects 

 33 

 
from other tissue types or from multiple fiber bundles with varying orientations, including fanning, 

crossing, or divergence, collectively referred to as crossing fibers. In a way, it is model-free because 

no underlying assumptions about microstructure are made, and the information about fiber 

orientation is entirely extracted from the response function estimated directly from the DWI signal.   

The major difference between previous CSD approaches and the algorithm implemented in 

MRtrix3’s software is a lack of normalization of the DWI signal, meaning the apparent fiber density 

metric is directly proportional to the measured signal.121 The metrics estimated in a fixel-based 

analysis (FBA) are the fiber density (FD), fiber cross-section (FC), and their product, the fiber 

density-cross section (FDC) (Table 1). FC compares the cross-sectional area perpendicular to a 

particular fiber orientation based on a subject’s warp to a study-specific template space. The 

multiplication of FD and FC to yield FDC creates a third and vital contrast for comparison in 

situations when a changing diameter accompanies a change in a fiber bundle’s axon quantity. Using 

these three measures, a group comparison of the degree to which a fiber population has had a 

decrease in density, a decrease in cross-section, or a simultaneous decrease in cross-sectional area 

and apparent number of axons is possible.   

A challenge uniquely addressed by FBA when modeling multiple fiber populations within a 

voxel is the reorientation of data for accurate spatial normalization before group comparison.137  The 

higher-order crossing fiber information has been shown to improve image registration between 

subjects, allowing for more accurate spatial alignment compared to FA registration.137 This allows 

for a comprehensive comparison of all WM fixels compared to traditional skeletonization techniques 

such as FSL’s Tract-based Spatial Statistics (TBSS) which reduces the analysis to a subset of WM 

voxels.138 Additionally, the estimation of fiber orientation distributions (FODs) using three tissue 

response functions and the associated reorientation techniques also help to address partial volume 

effects that skeletonization techniques like TBSS were developed to address.1  

 34 

 
Finally, the associated group analysis technique for FBA takes advantage of the improved 

tractography results using CSD estimated FODs to perform connectivity-based fixel 

enhancement.126,139 This builds on the threshold-free cluster enhancement proposed by Smith and 

colleagues, which focused on voxel enhancement based on their connection.140  The unique 

difference of the cluster-based fixel enhancement employed in FBA is the use of tractography from 

a study-specific template to inform the boosting of t-values. In this framework, a fixel that is 

connected via the tractography in a different location along a given tract is taken into account when 

boosting t-values. Thus, enhancement depends on membership to a tract as opposed to a local 

relationship in voxel space, which may not specifically define membership to a coherent WM tract 

and, by proxy, the projections of a specific cell population. 

Ultimately, NODDI and CSD are similar in that they both quantify the signal from the 

intraaxonal compartment to estimate the proportion of signal from inside axons and infer the 

density of axons in a volume. Additionally, all three methods, DKI, NODDI, and CSD, are capable 

of resolving aspects of crossing fiber architecture, increasing the specificity of these measures in 

brain regions containing complex microstructure.141 Each of these methods has been utilized in 

studying mTBI subjects, and fortunately, most studies also include DTI metrics. Moving forward, a 

thorough understanding of these techniques will be critical for mTBI researchers as they choose 

acquisition protocols and interpret the results of their diffusion studies using each technique. The 

robustness to crossing fibers of these methods and their increased specificity to the underlying 

pathophysiology in the WM will make them indispensable to traumatic brain injury researchers 

moving forward. 

 35 

 
 
Table 2.1 Common Diffusion Magnetic Resonance Imaging Metrics for DKI, NODDI, and FBA 

Metric 
(Abbr.) 

Mean 
Kurtosis 
(MK) 

Axial 
Kurtosis 
(AK) 

Radial 
Kurtosis 
(RK) 

Free Water 
Fraction 
(FWF/Viso) 

Neurite 
Density 
Index 
(NDI/Vic) 

Description 

A scalar summary measure calculated from the diffusion kurtosis tensor which is the 
average amount that the diffusion displacement probability distribution deviates from a 
normal or gaussian distribution.  

The average apparent diffusion kurtosis parallel to the principal diffusion 
direction/eigenvector. 

The average apparent diffusion kurtosis perpendicular to the principal diffusion 
direction/eigenvector. 

A measure of the contribution of the CSF signal within a voxel.  

A measure of the proportion of the intracellular signal within a voxel, which infers the 
density of axons within a voxel. 

Orientation 
Dispersion 
Index (ODI) 

Assesses the degree to which axons are traveling in a uniformly parallel orientation or 
dispersing away from one another in a fanning or scattering organization with a value close 
to 0 reflecting a uniformly parallel organization and 1 representing neurites dispersing in 
many directions. 

Fiber 
Density 
(FD) 

For a specific orientation of a fiber orientation distribution, FD represents an estimate of 
the amount of underlying axons.  

Fiber Cross 
Section (FC) 

An estimate of the diameter of a fiber population represented by a fixel based upon the 
warps calculated from registering a subject to a study-specific template image. 

Fiber 
Density 
Cross 
Section 
(FDC) 

The multiplication of the Fiber Cross Section and Fiber Density of a fixel which provides 
sensitivity to the situation where a fiber population is losing the number of axons within it 
in addition to shrinking in cross sectional areas.  

MATERIALS & METHODS 

PubMed was searched on June 12, 2023, and a second time on February 12, 2024. 

Three searches were conducted to compile studies that had utilized DKI, NODDI, or CSD and 

FBA. Keywords searched for mTBI were mild traumatic brain injury, concussion, mild brain trauma, 

or mild head injury, and for DKI were kurtosis, diffusion kurtosis imaging, DKI, or mean kurtosis 

 36 

 
 
  
tensor. The search was limited to papers published in the last 10 years. The DKI search yielded 22 

manuscripts abstracts which were reviewed for inclusion. One additional study outside this range 

was included due to its continuity with the publication by the same authors in 2013.142 Another DKI 

study was excluded because it was not available in English. Review articles, animal studies, 

interventional clinical trials, and manuscripts focused on methodologic development were excluded, 

leaving a total of 17 DKI manuscripts that are reviewed here. Keywords for NODDI were neurite 

orientation dispersion and density imaging, NODDI, water fractions, FISO, ODI, NDI, neurite 

density, and orientation dispersion. The NODDI search yielded 28 manuscripts whose abstracts 

were reviewed with the same criteria applied as in the DKI search, yielding 15 manuscripts. 

Keywords for CSD/FBA were constrained spherical deconvolution, fixel-based analysis, fixel, 

MRTrix3, CSD, and FBA. This search yielded 12 manuscripts whose abstracts were reviewed, with 7 

manuscripts included in this review. 

During the review of each article, key information about each sample’s characteristics, when 

available, was extracted, including injury mechanism (sports-related or general mechanisms from a 

population-based sample), data sample size, DWI data collection time points, neuroimaging software 

used for analysis and diffusion metrics estimated for each method DKI, NODDI, and FBA. In 

addition, the main findings of each manuscript, including the finding, comparison, timepoint, and 

localization of effects, were extracted.  

RESULTS & DISCUSSION 

Summary information from each reviewed manuscript is available by modeling technique for 

DKI (Table 4.2), NODDI (Table 4.3), and CSD/FBA (Table 4.4). 

Diffusion Kurtosis Imaging in mTBI 

One of the first studies to utilize DKI in a sample of mTBI patients during the acute period 

of recovery was by Karlsen & colleagues 2019. In this population-based sample from a Level 1 

 37 

 
 
trauma center and emergency clinic in Norway, decreased kurtosis fractional anisotropy (KFA) in 

mTBI patients compared to controls was observed. However, there were no other significant 

differences in kurtosis metrics. Interestingly, when comparing with DTI metrics, a greater number of 

significant voxels were observed with FA compared to KFA when comparing mTBI patients and 

controls. In contrast, later studies in similar populations in Norway by Stenberg & colleagues 2021 

observed the opposite relationship between FA and KFA in mTBI patients and controls. The work 

by Stenberg and colleagues found more significant voxels with lower KFA than FA between mTBI 

and controls. In this larger sample, a widespread decrease in all kurtosis metrics was seen at a 72-

hour time point during the acute phase of recovery (Table 2).143,144  In numerous comparisons 

between subgroups of mTBI patients, including participants with persistent post-concussive 

symptoms (PPCS) based on the British Columbia Post-concussion Symptom Inventory, reduced 

KFA was observed. Notably, significant differences were observed even when comparing mTBI 

patients with persistent symptoms at three months post-injury with mTBI patients who did not 

report PPCS. However, at the acute time point when analyses were controlled for cognitive reserve, 

no significant differences in any diffusion metrics remained. 

Contrasting patterns of DKI metrics have been observed, however, in contact sport athletes 

with sports-related mTBI. Two studies of male football players by Lancaster and colleagues 2016 & 

2018 demonstrated acute increases in axial kurtosis in widespread brain regions at 24-hour and 8-day 

time points following injury.145,146 No differences were observed for mean kurtosis or radial kurtosis. 

This study observed significant differences between groups at the 6-month time point in DTI 

metrics, including decreased MD, axonal diffusivity (AD), and RD. In more recent studies, elevated 

axial kurtosis was observed at 24-48 hours post-injury in two samples of male American football 

players.147,148 Radial kurtosis significantly decreased in the subacute phase at the 15-day timepoint in 

work by Muftler & colleagues and at a timepoint seven days after return to play following a sport-

 38 

 
related mTBI in work by Chung & colleagues.147,148 Increased KFA was observed in participants in 

frontal WM regions 15 days post-injury in both TBSS and voxel-based analysis (VBA) analyses. This 

finding was more pronounced at 45 days post-injury. However, a pattern of divergent KFA values in 

the control and mTBI groups was noted, indicating that observed differences may have been driven 

by decreasing KFA in the control group rather than changes in the mTBI group alone. Notably, 

these studies observed no between or within-group differences in DTI metrics in any comparisons, 

including VBA, TBSS, and ROI analyses.  

A later study from the same group conducted generalized linear mixed effects models to 

examine the relationship between prior concussion, years of contact sport participation, and 

DTI/DKI metrics.149 Interestingly, injury history was not associated with microstructural change, 

but years of contact sport exposure had a significant relationship with elevated RK. In the past, 

mTBI with a sports-related etiology was often discussed separately from other mTBI studies. 

However, this distinct divergence in DKI metrics in the acute phase of recovery between early 

population-based samples and contact sport samples highlights an opportunity to extract insight 

from these once separate areas of investigation. 

Many studies have been conducted in the sub-acute period following injury, with an early 

population-based sample recruited from a Maryland ED finding decreased mean and radial kurtosis 

in the internal capsule one month after injury.150 Although decreases in MK and RK at the 10-day 

post-injury time point were also observed in this study, this difference was not significant in the early 

subacute period. More recently, a population-based sample recruited from a neurosurgery clinic in 

China calculated MK between 2-weeks and 2-months of injury.151 This study found decreased MK in 

mTBI patients compared to controls in multiple regions of association fibers and projection fibers, 

and this was associated with lower digit span scores on a Digit Span Forward task. Still, another 

study observed decreased MK in thalamic, hippocampal, and callosal ROIs at 14-days and 3-months 

 39 

 
post-injury.152 The difference was significant only in the thalamic ROI from the first to second time 

point in this study of the subacute period following injury. A subsequent analysis of the same sample 

observed no differences when comparing ROI averaged DKI metrics between subsamples of 

patients with extensive symptoms on the Rivermead post-concussion questionnaire, and those with 

little to no symptoms.  

 In the chronic period, sports-related mTBI studies far outnumber population-based 

samples. Early work by Grossman and colleagues 2012 examined MK in a small population-based 

sample of 15 patients up to three years post-mTBI and found decreased MK in several brain regions 

when compared to sex, age, and education-matched controls (Table 2).142 A later study by the same 

group in 2013 found decreased MK in similar regions, including the thalamus and corpus callosum, 

over 9 months following injury.153 Finally, strong evidence from a large population-based sample of 

mTBI found sustained decreases in MK at a 12-month timepoint in widespread brain regions.144 

This suggests that decreases in MK that are observed in the acute and subacute time periods persist 

well beyond the typical recovery period for mTBI.   

Similarly, in the chronic period of recovery following injury, a decrease in MK is commonly 

reported at 6 months post-injury150 and was even observed in two samples at post-season scans after 

a season of high school football in players without mTBI.154,155 Still, other studies have observed no 

significant differences in some or all DKI metrics at 6 months post-injury in football players.145,148 

Nonetheless, it is clear that in some populations, an incompletely understood process of change in 

WM may persist for months to years following a traumatic brain injury, even those presently 

categorized as mild.  

 The question of what differences exist in repetitive head injury, and subconcussive blows 

have also been investigated with DKI. An interesting study by Gong & colleagues compared pre and 

post-season scans in high school athletes who were monitored during practice and games with head 

 40 

 
 
 
impact telemetry during a football season.154 They found significant decreases in MK in cortical and 

deep WM, including the thalamus, over the course of the season. They also found that, although also 

significant, the DTI summary metric MD had a smaller effect size in the thalamus when compared 

to MK. Another similar study of high school football players without mTBI by Davenport & 

colleagues utilized DKI in pre and post-season scans.155 They found a significant association of risk-

weighted cumulative exposure combined probability, a summary measure of risk associated with 

linear and rotational acceleration experienced during head blows, with voxels >2 standard deviations 

below the mean in DKI metrics MK, AK, and RK. These studies provide early evidence that caution 

must be exercised in interpreting WM changes in the context of sports-related mTBI. These 

differences may not necessarily be due to a singular impact that causes a diagnosed mTBI, but rather 

demonstrate the unique aspects of this mTBI etiology in the study of WM response to repeated 

traumatic insults in contact sport athletes.  

Overall, studies of mTBI utilizing DKI observed mixed patterns of WM change with 

kurtosis metrics in the acute period following injury. In the acute period, populations of athletes, 

mostly football players, had increased MK148,156, RK156, and AK146–148 compared to controls. In 

contrast, population-based studies observed decreases in MK143,144, RK143, and AK143. In population-

based samples in the subacute period, the majority of studies reported decreased MK150,151,153 , with 

some observing continued decreases in KFA157, and RK150.  One study reported an increase in 

MK152, and one reported no differences in DKI metrics during the subacute period.144 Only one 

study of sport-related mTBI examined group differences in DKI metrics in the subacute period and 

noted increased MK, decreased RK, and increased KFA.147 In the chronic period following injury, 

sport-related and population-based studies most commonly observed decreased MK.142,144,150,153 Most 

interestingly, studies collecting pre & post-season scans in otherwise healthy athletes observed 

decreased MK at post-season scans.154,155 

 41 

 
Neuroimaging 
software & 
metrics 
DESIGNER 
software for DKI 
analysis. MRTrix3 
and automated 
fiber 
quantification 
(AFQ) for 
tractometry. 

Metrics: MK, AK, 
RK 

Significant metrics and 
localization of effects 

Increased MK and RK 
in the L UF and R cing 
when comparing 
preseason scans with 24-
48 hr scans.  
Increased MK in the R 
cing cingulate in football 
players who had a prior 
concussion or sustained 
an injury in the study 
when compared with 
volleyball players. 

Article 

Goubran et 
al., 2023 

Table 2.2 Summary of DKI in mTBI Studies 
Comparison 
group 
demographics 
Football 
Players:  
n = 63 (0F, 
63M) 
Age: mean ± 
SD 
(19.11±1.57) 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
Annual pre-
season scan.  

Timepoint 2:  
Within 24-48 
hours of injury 
mean ± SD 
(1.83d±1.26) 

Volleyball 
Players: 
n = 34 (0F, 
34M) 
Age: 
(19.57±0.91) 

Timepoint 3:  
6 mo followup 
scan 
(165d ± 77) 

Timepoint 4:  
After final 
athletic season 

 42 

 
 
 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Stenberg et 
al., 2023 

Comparison 
group 
demographics 
Population 
Sample mTBI:  
n = 193 (70F, 
123M) 
Age: median, 
range (27, 16-
59) 

mTBI 
Subsample 
Complicated 
mTBI:  
n = 22   

Age, gender, 
and education-
matched 
community 
controls:  
n = 83 (33F, 
50M) 
Age: (27.7, 16-
58) 

Neuroimaging 
software & 
metrics 
Diffusional 
Kurtosis 
Estimator, TBSS-
based voxel-wise 
analysis (FSL) 

Metrics: KFA, 
MK, AK and RK  

Post-injury 
measurement 
timepoints 
Timepoint 1: 
72h  
mean ± SD 
(52h±19h) 

Timepoint 2: 
3mo  
(95d±7d) 

Timepoint 3: 
12mo 
(370d±17d)  

Significant metrics and 
localization of effects 

A longitudinal analysis 
of diffusion metrics 
revealed that mean 
kurtosis was relatively 
stable over time in the 
mTBI group. 
Fluctuations in the 
control group drove 
interaction effects.  
Decreased MK in the 
CR, CC, cing, IC, Fx, 
TR, CST, sagittal 
stratum, cerebellar 
peduncle, and medial 
lemniscus at 72h time 
point.   
Decreased MK in the 
CR, CC, cing, IC and 
EC, Fx, TR, SLF, and 
sagittal stratum at 12 mo 
time point. 

 43 

 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Chung et al., 
2022 

Comparison 
group 
demographics 
Collegiate 
athletes mTBI: 
mTBI 
subsample: 
football 
players 48 hr 
post   
n = 24 (0F, 
24M) 
Age: mean ± 
SD (19.7 ± 
1.1) 

mTBI 
Subsample: 
Repeat head 
injury (RHI) 
with no mTBI 
during study: 
n = 26 (0F, 
26M)  
Age: (19.3 ± 
1.2) 

Non-contact 
sport controls 
w/out head 
injury:  
n = 28 (0F, 
28M) 
Age: (19.7 ± 
1.4) 

Neuroimaging 
software & 
metrics 
In-house 
MATLAB scripts 
to estimate DKI 
metrics, TBSS-
based voxelwise 
analysis . Post-
hoc ROI analysis 
using ICBM-DTI-
81. 

Metrics: MK, AK, 
RK 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
24-48h post 
injury  

Timepoint 2:  
Asymptomatic 
period after 
return to play 
clearance  
mean ± SD 
(8.1d±5.6d) 

Timepoint 3:  
7 days after 
return to play 
(27d ± 12.5d) 

Timepoint 4:  
6-months post 
injury 
(182d ± 14d) 

Significant metrics and 
localization of effects 

Increased AK in mTBI 
group compared to 
controls.  
The extent of 
significantly different 
voxels for AK decreased 
by the 6-month 
timepoint when 
comparing mTBI and 
controls. 
Increased MK in mTBI 
compared to controls in 
the EC, posterior limb 
and retrolenticular part 
IC, cerebral peduncle, 
UF, ILF, posterior TR 
and splenium of the CC.  
Differences in MK did 
not persist at the 6 mo 
time point. 
Decreased RK in mTBI 
compared to controls 7 
days after return to play 
I the posterior CC.  
Increased AK and MK 
in the RHI group 
compared to controls in 
the CC, CR. SLF, 
posterior limb and 
reticulolenticular part of 
IC, posterior TR and 
cerebral peduncle.  
No significant 
differences between 
mTBI and RHI groups. 

 44 

 
 
 
 
 
 
 
 
Significant metrics and 
localization of effects 

Decreased MK in mTBI 
patients compared to 
controls, with no 
differences in FA or 
MD in the bilateral SLF, 
cing, R ILF, IFF, UF, 
inter-hemispheric fibers 
of body of CC, 
projection fibers of 
CST, ATR, and bilateral 
IC and EC.  
Attention deficit as 
measured by lower digit 
span forward score was 
significantly associated 
with decreased MK in 
the mTBI group in the 
SLF, ILF, IFF, UF, 
body of the CC, CST, 
ATR, bilateral IC and 
EC.  
Decreased MK in mTBI 
patients in the right 
hippocampus, left 
thalamus, left caudate, 
right putamen, and right 
pallidum.  
History of concussion 
was not independently 
associated with 
microstructural or 
macrostructural WM 
changes.  

Significant associatons 
between years of self-
reported contact sport 
exposure and WM 
microstructural 
abnormalities. 

Table 2.2 (cont’d)  
Article 

Wang et al., 
2022 

Comparison 
group 
demographics 
Population 
Sample mTBI:  
n= 23 (11F, 
12M) 
Age: mean ± 
SD (47± 9.9) 

Post-injury 
measurement 
timepoints 
Timepoint:  
2wk – 2mo  
mean (39d) 

Neuroimaging 
software & 
metrics 
Diffusional 
Kurtosis 
Estimator, TBSS-
based voxel-wise 
analysis (FSL) 

Metrics: MK 

Controls:  
n = 24 (12F, 
12M) 
Age: (49± 
13.5) 

Brett et al., 
2021 

Timepoint 1:  
48h 

Timepoint 2: 
8d  

Timepoint 3:  
15d  

Timepoint 4: 
45d  

Football 
Players & 
Non-contact 
sport players:  
n = 121 (0F, 
121M) 
Age: mean ± 
SD 
(18.52±1.74) 

Years of 
contact sport 
exposure 
(5.01y ± 4.49y) 

Diffusional 
Kurtosis 
Estimator, TBSS-
based voxel-wise 
analysis (FSL). 

Generalized linear 
mixed effects 
models examining 
relationship of 
prior concussion, 
years of contact-
sport 
participation and 
WM 
microstructure. 

Metrics: KFA, 
MK, AK and RK 

 45 

 
 
 
 
 
 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Stenberg et 
al., 2021 

Comparison 
group 
demographics 
Population 
Sample mTBI:  
n= 176 (65F, 
111M) 
Age: median, 
IQR (28.1, 22) 

mTBI 
Subsample 
Complicated 
mTBI: n = 18   

Age, gender, 
and education-
matched 
community 
controls: 
n = 78 (30F, 
48M) 
Age: (27.6, 20) 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
72h 
mean ± SD 
(52h±19h) 

Neuroimaging 
software & 
metrics 
Diffusional 
Kurtosis 
Estimator, TBSS-
based voxel-wise 
analysis (FSL) 

Metrics: KFA, 
MK, AK and RK 

Significant metrics and 
localization of effects 

Decreased KFA and 
higher RD in PPCS 
patients in the CC, CR, 
IC, and TR than patients 
without PPCS.  
No significant 
differences remained in 
any diffusion metrics 
between patients with 
and without PPCS when 
controlling for cognitive 
reserve. 
Decreased KFA, MK, 
AK, RK in the CC, CR, 
IC, SLF, and TR. 
Decreased MK was also 
present in the 
cerebellum and 
brainstem. 
These differences 
remained when 
examining only the 
uncomplicated mTBI 
patients. 
Decreased MK, AK and 
RK in mTBI patients 
without PPCS compared 
to controls in the IC, 
cerebellum, brainstem, 
and thalamus. 

 46 

 
 
 
 
Significant metrics and 
localization of effects 

Increased AK in mTBI 
compared to controls 
with increasing number 
of significant voxels 
from 48h to 15d 
timepoint with no 
significant differences at 
the 45d timepoint.  
Decreased RK in mTBI 
compard to controls at 
the 15d timepoint with 
both the whole brain 
voxel-wise and TBSS 
analysis, and at the 45d 
timepoint for the whole 
brain voxelwise analysis 
only.  
Increased KFA in mTBI 
compared to controls in 
frontal WM regions at 
th 15d and 45d 
timepoint.  
Decreased KFA in 
mTBI compared to 
controls in the genu of 
the CC, cerebellar 
peduncle, IC, CR, and 
SFF at the 72h 
timepoint.  
Five times the voxels of 
decreased FA and KFA 
were observed at the 
3mo timepoint in the 
same regions with the 
addition of the posterior 
TR bilaterally.  
No significant 
differences for MK, AK 
or RK.  

Table 2.2 (cont’d)  
Article 

Muftler et al., 
2020 

Comparison 
group 
demographics 
Football 
Players: n= 96 
(0F, 96M) 
Age: mean± 
SD (18.06 ± 
1.5) 

Controls:  
n = 82 (0F, 
82M) 
Age: (18.37 ± 
1.7) 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
48h 
mean ± SD 
(32.47h ± 14.19) 

Timepoint 2: 
8d  
(8.20d ± 0.98) 

Timepoint 3:  
15d  
(15.42d ± 1.35) 

Timepoint 4: 
45d  
(45.56d ± 3.77) 

Neuroimaging 
software & 
metrics 
Diffusional 
Kurtosis 
Estimator, TBSS-
based voxel-wise 
analysis (FSL), 
Whole-brain 
Voxel Based 
Analysis, 
Longitudinal ROI 
analysis based on 
significant 
clusters from 1st 
level TBSS 
analysis. 

Metrics: KFA, 
MK, AK and RK 

Karlsen et al., 
2019 

Population 
Sample mTBI:  
n = 25 (13F, 
12M) 
Age: mean ± 
SD (32.7 ± 
13.0) 

Timepoint 1: 
72h 
mean, range 
(69h, 261h)  

Timepoint 2: 
3mo 
(82d, 43d) 

Controls:  
n = 22 (12F, 
10M) 
Age: (34.5 ± 
8.7) 

TBSS-based 
voxel-wise 
analysis (FSL) 

5 WM ROIs 
created with 
intersection of 
ICBM-DTI-81 
and the TBSS 
WM skeleton 

Hand drawn 
thalamic ROIs 

Metrics: KFA, 
MK, AK and RK 

 47 

 
 
 
 
 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Gong et al., 
2018 

Lancaster et 
al., 2018 

Comparison 
group 
demographics 
Football 
Players w/out 
mTBI: 
n = 16 (0F, 
16M) 
Age:  median, 
range 
(16y, 15-17) 

Years of 
contact sport 
exposure 
(7y, 5-12) 

Football 
Players:  
mTBI:  
n=17 (0F, 
17M) 
Age: 
mean±SD 
(17.5± 1.7) 

Age, gender, 
sport, 
premorbid 
level of verbal 
intellectual 
functioning, 
and GPA 
matched 
controls:  
n = 20 (0F, 
20M) 
Age: 
(17.9±1.7) 

Significant metrics and 
localization of effects 

Decrease in MK in the 
thalamus, left 
paracentral, right pars 
triangularis, right 
inferior parietal, right 
cuneus, and right rostral 
middle frontal cortices 
from the pre- to post-
season scan.  

A larger effect size was 
observed for pre to 
post-season differences 
in MK in the thalamus 
suggesting it may be 
more sensitive than MD. 
No differences in DKI 
metrics at the six mo 
timepoint or in 
longitudinal analyses 
when comparing 
controls and mTBI 
patients.  

Neuroimaging 
software & 
metrics 
In-house 
MATLAB scripts 
to estimate DKI 
metrics. 

Metrics: MK 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
Pre-season 
before first 
contact practice 
Mean, range 
(2d, 2-6d) 

Timepoint 2:  
Post-season  
(10d, 2-6d) 

Timepoint 1: 
24hr  
mean, range 
(19.69hr, 14-
24hr) 

Timepoint 2: 
8d  
(8.3d,  7-11d) 

Timepoint 3: 
6mo 
(168.56d,  151-
204d) 

TBSS-based 
voxel-wise 
analysis (FSL) 

DTI tensors and 
DKTI tensors 
were estimated 
using in house 
software. 

ROI based 
analysis of WM 
tracts based on 
the ICBM-DTI-
81 WM atlas at 
the 6mo 
timepoint. 

Metrics: MK, AK, 
RK 

 48 

 
 
 
 
 
 
 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Naess-
Schmidt et al, 
2018 

Naess-
Schmidt et 
al., 2017 

Comparison 
group 
demographics 
Population 
Sample: 
mTBI w/ 
RPQ ≥ 20  
 n= 25 (16F, 
9M,) 
Age: mean± 
SD (24± 3.9) 

mTBI w/ 
RPQ = 0 
n= 25 (11F, 
14M) 
Age: mean± 
SD (22.7± 3.5) 

Controls:  
n = 27 (16F, 
11M) 
Age: (27 ±  
6.2) 
Population 
Sample mTBI:  
 n= 27 (16F, 
11M) 
Age: mean± 
SD (27.6± 6.4) 

Controls:  
n = 27 (16F, 
11M) 
Age: (27.4 ±  
6.2) 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
2-5mo post 
mTBI 

Neuroimaging 
software & 
metrics 
Combination of 
FSL and inhouse 
MATLAB scripts 
to estimate MKT. 
Automatic 
segmentation of 
ROIs. 

Metrics: MKT 

Significant metrics and 
localization of effects 

No significant 
differences in diffusion 
metrics in any ROIs 
between patients with 
extensive symptoms on 
RPQ vs. those with 
minimal symptoms.  

Increased MK in mTBI 
compared to controls in 
the thalamus at the 14d.  
Decreasing MK in 
mTBI thalamic ROIs 
when comparing within 
group between the 14d 
and 3 mo timepoint. 

Timepoint 1: 
14d 
mean± SD 
(10.3d ± 3) 

Timepoint 1: 
3mo 
(100.1d ± 7) 

Combination of 
FSL and inhouse 
MATLAB scripts 
were used to co-
register auto-
segmented 
thalamic, 
hippocampal and 
callosal ROIs 
from 
MP2PRAGE 
images with 
corresponding 
regional mean 
images. 

Metrics: MK 

 49 

 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Davenport et 
al., 2016 

Lancaster et 
al., 2016 

Comparison 
group 
demographics 
Football 
Players w/no 
mTBI:  
n=24 (0F, 
24M) 
Age: mean± 
SD (16.9 ± 
0.6) 

Football 
Players:  
mTBI:  
n=26 (0F, 
26M) 
Age: mean± 
SD (17.6 ± 
1.5) 

Age, gender, 
sport, 
premorbid 
level of verbal 
intellectual 
functioning, 
and GPA 
matched 
controls:  
n= 26 (0F, 
26M) 
Age: (18 ± 1.5) 

Post-injury 
measurement 
timepoints 
Timepoint 1:  
Baseline 

Timepoint 2:  
Pre-season 

Timepoint 3:  
Post-season 

Interval 
between pre & 
post season 
scans: mean± 
SD 
(4.9mo ± 0.6) 
Timepoint 1: 
24hr  

Timepoint 2: 
8d 

Neuroimaging 
software & 
metrics 
Diffusional 
Kurtosis 
Estimator, FSL & 
SPM 

Metrics: MK, AK, 
RK 

Significant metrics and 
localization of effects 

A significant association 
between risk weighted 
cumulative exposure 
combined probability 
and voxels >2 standard 
deviations below the 
mean for MK, AK and 
RK.  

TBSS-based 
voxel-wise 
analysis (FSL) 

DTI tensors and 
DKTI tensors 
were estimated 
using in house 
software. 

Metrics: MK, AK, 
RK 

Increased AK in the IC, 
UF, cerebral peduncle, 
CST, IFF, and ILF 
when comparing 
controls and mTBI at 
24hr time point.  
There were more 
widespread regions with 
increased AK at the 8d 
time point.  
No observed differences 
for MK or RK.  

 50 

 
 
 
 
 
 
 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Stokum et al., 
2015 

Grossman et 
al., 2013 

Comparison 
group 
demographics 
Population 
Sample mTBI:  
n= 24 (6F, 
18M) 
Age: mean± 
SD (37.4± 
14.1) 

Controls:  
n = 24 (11F , 
13M) 
Age: (33.9 ± 
14.7) 
Population 
Sample mTBI:  
n= 20 (4F, 
16M) 
Age: mean± 
SD (34.8± 
10.7) 

Sex, age, and 
education 
matched 
controls:  
n= 16 (3F, 
13M) 
Age: (35.1 ± 
11.9) 

Post-injury 
measurement 
timepoints 
Timepoint 1: 
10d 
mean± SD (6 ± 
3d) 

Timepoint 2: 
1mo 
(33 ± 7d) 

Timepoint 3: 
6mo 
(196 ± 33d) 

Timepoint 1: 
1mo  
mean± SD 
(22.1d ± 15.4) 

Timepoint 2: 
>9mo 
(369.6d ± 112.1) 

Neuroimaging 
software & 
metrics 
Custom hand 
drawn ROIs and 
in house 
MATLAB scripts 
utilized to analyze 
mean and 
standard 
deviation of 
diffusion metrics 
in each ROI.  

Metrics: MK, RK, 
lr 

In-house 
MATLAB scripts, 
TBSS-based 
voxel-wise 
analysis (FSL) 

Metrics: MK 

Significant metrics and 
localization of effects 

Decreased RK and MK 
were observed in the 
anterior limb of the IC 
at the 1mo and 6mo 
timepoint.  
No differences in lr 
when comparing mTBI 
patients and controls at 
each timepoint.  

Decreased MK in mTBI 
compared to controls in 
the thalamus, IC, EC, 
CC, cing, optic 
radiations, centrum 
semiovale, total deep 
gray matter, and total 
WM at the 1mo time 
point. 
In the thalamus DTI, 
DKI and ASL metrics 
were significantly 
associated.  
Decreased MK in mTBI 
patients in the thalamus, 
CC, cing, optic 
radiations, centrum 
semiovale, total deep 
gray matter, and total 
WM at the >9 month 
timepoint. 

 51 

 
 
 
 
 
 
 
 
 
Table 2.2 (cont’d)  
Article 

Grossman et 
al., 2012 

Comparison 
group 
demographics 
Population 
Sample mTBI:  
n= 22 (8F, 
14M) 
Age: mean± 
SD (38.2± 
11.7) 

Sex, age, and 
education 
matched 
controls:  
n= 14 (5F, 
9M) 
Age: (36.5 ±  
12.3) 

Neuroimaging 
software & 
metrics 
In-house 
MATLAB scripts, 
SPM, and ImageJ.  

Hand-drawn 
ROIs with 
average diffusion 
metrics were 
compared for the 
thalamus, IC, CC, 
and centrum 
semiovale.  

Significant metrics and 
localization of effects 

Decreased MK in mTBI 
patients compared to 
controls.  
Decreased MK in 
Group 1 patients 
compared to controls in 
the thalamus.  
Decreased MK in 
Group 2 patients 
compared to controls in 
the thalamus, IC, 
splenium of the CC and 
centrum semiovale.  

Post-injury 
measurement 
timepoints 
Follow up 
varied by group:  

Group 1 (7 
mTBI patients) 

Timepoint 1: 
Within 1 year of 
injury 
Mean, range 
(65.7d, 14.6-
215.35d)  

Group 2 (15 
mTBI patients)  

Timepoint 1: 
>1 year after 
injury 
(3.9y, 1.33-
9.58y) 

NODDI in mTBI 

Early work with NODDI by Churchill and colleagues in a sample of athletes with 

sports-related concussions revealed correspondence between acute increases in RD and decreases in 

VIC.158 No difference in VISO or ODI was observed, suggesting that these findings were not related to 

increases in free water or a change in the organization of fibers within colocalized regions of 

significant RD and VIC in this sample. Only sparse areas of longitudinal change were observed in 

ODI, and other DTI metrics preceding participants' return to play, seemingly suggesting that during 

the recovery period in some mTBI patients, significant changes in WM microstructure may not 

coincide with symptomatic recovery and current clinical definitions of recovery.  

 52 

 
 
 
 
 
 
 
 
 
 
 
Although not a purely mTBI sample, a study of repetitive mTBI in mixed-martial arts 

athletes scanned during a professional competition season similarly demonstrates the relationship 

between NODDI and DTI.159 In this sample, decreases in FA were colocalized with increases in 

ODI, demonstrating the inverse relationship between these two metrics. Similarly, decreased MD 

was related to decreased VISO, while VIC did not demonstrate significant overlap with either FA or 

MD. Estimates of variance were calculated for the relationship between NODDI metrics as 

explanatory variables and DTI metrics as dependent variables, and it was demonstrated that only 

moderate relationships exist between Vic, Viso, and DTI metrics. Taken together, this study supports 

the notion that although complementary, the biophysical model NODDI offers increased specificity 

for pathologic microstructural change beyond DTI.  

In a study of (n =32) CT/MRI negative mTBI patients, patients’ DTI, DKI, and NODDI 

metrics from a multi-shell DWI were compared in the acute phase of injury with a median scan time 

of 29 hours.160 TBSS whole-brain VBA revealed numerous regions of significant difference between 

mTBI patients and controls. The largest areas of difference by significant voxel number were with 

AD in the right posterior thalamic radiation, external capsule, and internal capsule. Similarly, for 

NODDI metrics, large effects for ODI were observed bilaterally in the posterior thalamic radiation, 

posterior limb of the internal capsule, and external capsule. In ROC analysis of whole brain WM 

averages ODI had the best performance differentiating mTBI and control patients compared to all 

other metrics (AUC = 0.73).  

Some studies utilize hybrid analysis methods, such as the work by Oehr and colleagues 2021, 

in which an ROI analysis was conducted using average DWI metrics calculated using hand-drawn 

tractography masks.161 In this study, the NODDI metrics ODI, ICVF, and ISO were compared 

between mTBI patients and orthopedic trauma controls between 6 to 12 weeks following injury. 

The NODDI metric ODI was significantly higher in this sample of mTBI patients when compared 

 53 

 
 
with controls in the left uncinate fasciculus. ISO also demonstrated significant group-level 

differences, with the measure being lower in the mTBI group in the corpus callosum, and bilaterally 

in the superior longitudinal fasciculus direct and indirect segments. Interestingly, no differences were 

observed between mTBI patients and controls for the NODDI metric ICVF or the DTI metrics 

MD, AD, and RD. This suggests that edema related to the injury is resolved 6-12 weeks post-injury 

in this sample. The observation of reduced FA without concurrent changes in other DTI metrics is 

difficult to explain; however, it may be due to the smaller sample size of this study. Also, averaging 

metrics across entire WM tracts can be problematic, as the analysis technique can mask effects in 

certain more susceptible regions along the tract. Nonetheless, from a clinical perspective, this is 

likely a step closer to an approach that could be useful in single patients to gain a rough idea of these 

metrics in comparison to population averages in specific WM tracts.   

An analysis of DWI data collected in pediatric mTBI patients in the Advancing Concussion 

Assessment in Pediatrics (A-CAP), revealed no difference in any diffusion metrics.162 A possible 

explanation for these results could be the low number of directions collected for the b=2000 

direction, as the optimized protocol for NODDI calls for 60 directions at b=2000. Recent work by 

Seider and colleagues suggests that the number of directions needed to fit diffusion models 

accurately could be significantly greater than was previously thought.163  Additionally and similar to 

other studies, due to the heterogeneous and mild nature of the injury, tract averaging may obscure 

the ability to localize individual differences in specific locations along a tract.  

In contrast, results from the PLAYGAME trial found decreased free water fraction in grey 

matter regions often colocalized with reduced MD, while differences between groups with mTBI 

having increased intracellular water fraction were generally more widespread and did not overlap 

with DTI metrics.164 This study used a modified approach to calculate NODDI metrics without the 

fixed rates of intra and extracellular diffusivity in the original model for a more biologically informed 

 54 

 
 
multicompartmental diffusion model with tissue-specific rates of diffusivity.165 They noted that there 

were different statistical relationships between traditional NODDI metrics and those calculated with 

a more biologically plausible tissue-dependent diffusivity value.     

Further evidence for NODDI metrics’ sensitivity in pediatric mTBI comes from another 

study that sought to test NODDI metrics in outcome prediction models with multiple logistic 

regression.166 In this study, they found that the model incorporating global WM ROI ODI at 1-

month post-injury had the greatest performance in predicting the 2-3 month outcome with 81.82% 

and 87.18% negative and positive predictive power, respectively. This significantly outperformed the 

null clinical predictor model, with 58.8% and 70.5% negative and positive predictive power, 

respectively. Interestingly, models including FISO did not outperform the null models.  

Currently, there are no interventions to improve outcomes following mTBI, so an 

understanding of factors following injury that may affect recovery is important. In a recent study 

examining the interaction of self-reported sleep quality in the first week following injury, it was 

noted that adolescent mTBI patients with poor sleep quality had decreased NDI in almost all tracts 

investigated when compared to those with good sleep quality and healthy controls.167  

Another development in the field is the use of novel acquisition sequences, such as Hybrid 

diffusion imaging (HYDI), that would allow the application of numerous signal modeling 

techniques. A recent study of mTBI patients with chronic symptoms 3 months post-injury utilizing a 

HYDI acquisition found significant differences in numerous WM regions in the NODDI metrics 

ODI and Vic.168 No differences in DTI metrics were observed in this study. In another study using 

HYDI acquisition, mTBI patients had lower Vic compared to controls in the corpus callosum and 

corona radiata.122 This study also correlated diffusion metrics with neuropsychological performance 

on the Delis–Kaplan Executive Function System and found that increasing Vic was associated with 

decreased performance in the form of increased reaction time and decreased recall on short and 

 55 

 
long-delayed recall tasks. This would seem to suggest these higher-order modeling techniques are 

more sensitive to acute and chronic WM change, even using a clinically feasible acquisition that takes 

only 8 minutes to acquire. More recently, work by Anderson and colleagues demonstrated a lack of 

relationship between NODDI metrics and measures of cognitive performance that was present in 

mTBI patients but not control subjects.169 Taken together, these studies demonstrate that not only is 

this higher-order modeling technique sensitive to changes in WM but also specific for differences in 

neuropsychological tests. This will be important in understanding how these WM changes relate to 

altered function. 

Additional strong evidence that NODDI is more sensitive in detecting changes in WM 

microstructure than DTI comes from the analysis of data in the TRACK-TBI study.112 A 

comprehensive analysis of patients, friend controls, and orthopedic injury controls was collected 

during the initial study, and a second replication sample of mTBI patients was conducted on multi-

shell DWI data at b=1300 mm/s2 and 3000 mm/s2. Higher FISO was observed in the mTBI group 

compared to controls at the 2-week timepoint in widespread brain regions, including the genu and 

body of the corpus callosum, anterior and posterior limbs of the internal capsule, anterior corona 

radiata, anterior thalamic radiation, external capsule, cingulum, superior longitudinal fasciculi, 

posterior corona radiata, and inferior frontal-occipital fasciculus. Conversely, lower NDI in mTBI 

patients compared to controls was observed at two weeks post-injury in the external capsule, 

anterior thalamic radiation, inferior longitudinal fasciculi, fornix, and stria terminalis. Significant 

longitudinal decreases in NDI were observed between the 2-week and 6-month timepoints in the 

anterior corona radiata, posterior corona radiata, posterior thalamic radiation, inferior longitudinal 

fasciculi, inferior frontal-occipital fasciculus, anterior thalamic radiation, external capsule, and 

uncinate fasciculi. Interestingly, increases in FISO among mTBI patients was associated with 

membership to a cluster with greater improvement on a composite score for recovery in an 

 56 

 
unsupervised machine learning analysis. These observations were made in the context of either no 

difference or less extensive voxel-wise differences in DTI metrics.  This would seem to support the 

notion that higher-order models are more sensitive and, therefore, more useful in characterizing the 

subtle WM changes thought to be driving symptoms following mTBI.  

Finally, another study examined the association between peripheral blood biomarkers 

commonly associated with aspects of neurotrauma (tau, neurofilament light (NfL), and glial fibrillary 

acidic protein (GFAP)), and NODDI metrics.170 In this study, healthy high school football players 

with a history of a prior football season in the past 12 months had serum collection and DWI scans 

at a preseason baseline timepoint. Modest associations were observed including a negative 

association between serum tau and NDI, and a negative association between NfL and NDI. 

Notably, MD had widespread positive associations with plasma tau in this analysis.  

Contemporary studies will benefit from utilizing multiple signal representation and higher-order 

modeling techniques. For example, Goubran & colleagues compared DTI, NODDI, and DKI 

metrics in a longitudinal sample of male college football and volleyball athletes (Tables 2 & 3). The 

longitudinal comparison of time and sport, which excluded post-injury scans, demonstrated a 

pattern of increasing FA in volleyball players and decreasing MD/RD and ODI throughout the 

season when compared to football players. This adds further strong evidence that changes observed 

in non-injured subjects may drive the statistical differences in diffusion metrics via disruption of 

healthy WM remodeling. However, a subsequent analysis demonstrated an increase over time in 

ODI mediated by player position, casting uncertainty on the possibility of altered WM dynamics as a 

possible mechanism. This increase in WM dispersion and lack of change in other NODDI metrics 

demonstrate the real possibility that dynamic partial volume effects in mTBI are a source of 

unaccounted variance in DKI and DTI studies. Additionally, in contact sport samples, the degree 

and type of impact may play more of a role in observed changes than previously understood. 

 57 

 
Neuroimaging 
software & 
metrics 
MRTrix3, FSL, 
NODDI 
MATLAB 
toolbox 

Metrics: ODI, 
ICVF 

DESIGNER 
software for 
DKI analysis. 
MRTrix3 and 
automated 
fiber 
quantification 
(AFQ) for 
tractometry. 

Metrics: Fiso, 
ODI and Ficvf 

Significant metrics and 
localization of effects 

ICVF was positively 
predictive of processing 
speed ability and memory 
index performance for the 
control group in 
numerous WM tracts. 

ICVF was negatively 
predictive of executive 
function index in the SLF 
only.  
ODI decreased 
longitudinally in volleyball 
players but not football 
players in the forceps 
minor of CC, L superior 
SLF, L TR, R cing 
ODI decreased 
longitudinally in non-
injured football players 
compared to injured.  
ODI increased in players 
with high position based 
impact risk over the 
course of the season.  

Table 2.3 Summary of Studies Utilizing NODDI in mTBI Patients 
Article  

Anderson 
et al., 2023 

Goubran 
et al., 2023 

Comparison 
group 
demographics 
Hospitalized 
mTBI:  
n = 26 (4F,22M) 
Age: mean ± SD 
(34.81 ± 13.76) 

Post-injury 
Measurement 
Timepoints 
Timepoint 1:  
6-12 wks post 
injury 
mTBI mean ± 
SD 
(63.269 ± 12.127) 

Trauma controls:  
n =  20 (2F, 18M) 
(38.75± 12.59) 

Controls:  
(49.95 ± 9.185) 

Football Players: 
n = 63 (0F, 63M) 
Age: mean ± SD 
(19.11±1.57) 

Volleyball Players:  
n = 34 (0F, 34M) 
Age:  
(19.57±0.91) 

Timepoint 1: 
Annual pre-
season scan.  

Timepoint 2: 
Within 24-48 
hours of injury 
mean ± SD 
(1.83d± 1.26) 

Timepoint 3: 6 
month followup 
scan 
(65d ± 77) 

Timepoint 4: 
After final 
athletic season 

 58 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 2.3 (cont’d)  
Article  

Stein et al., 
2023 

Huang et 
al., 2022 

Comparison 
group 
demographics 
Pediatric 
population 
sample mTBI: 
Symptomatic 
group  
n = 80 (46F, 
34M) 
Age: median, IQR 
(14.6y, 3.2) 

Asymptomatic 
group:  
n = 32 (12F, 
20M) 
Age: (14.5y, 3.1) 

Controls: n = 21 
(12F, 9M) 
Age: (14.7y, 6.8) 

Patients:  
mTBI:  
n= 32 (15F, 17M) 
Age: mean ± SD 
(32 ± 13)  

Controls:   
n = 31 (17F, 
14M) 
Age: (37 ± 9) 

Significant metrics and 
localization of effects 

Increased ODI in 
symptomatic mTBI 
compared to all other 
groups in the global WM, 
bilateral UF and bilateral 
IFF at both 1 mo and 2-
3mo time points. 

Decreased ODI, Vic and 
Viso were significantly 
lower throughout the 
brain, especially in long-
association fiber and 
commissural fiber tracts.  

Post-injury 
Measurement 
Timepoints 
Timepoint 1: 1-
month post 
injury  
mean ± SD 
(37.9d± 5.7) 

Timepoint 2:  
2-3 months post 
injury 
(69.3± 6.3) 

Timepoint 1: 
Median: 29 hours 
post-injury  

Neuroimaging 
software & 
metrics 
Voxel-wise 
analysis with 
the 
Multivariate 
and Repeated 
Measures 
(MRM) 
toolbox. ROI 
analysis based 
on significant 
voxel-wise 
statistical map. 

Metrics: ODI, 
FISO 

TBSS-based 
voxel-wise 
analysis (FSL) 
ROC analysis 
to discriminate 
mTBI from 
controls with 
whole brain 
WM averages 

Metrics: ODI, 
Vic and Viso 

 59 

 
 
 
 
 
 
 
 
 
Table 2.3 (cont’d)  
Article  

Mayer et 
al, 2022 

Comparison 
group 
demographics 
Pediatric 
population 
sample mTBI:  
n= 204 (83F, 
121M) 
Age: mean ± SD 
(14.5y ± 2.9)  

Controls:  
n = 173 (73F, 
100M) 
Age: (14.2y ± 2.8) 

Post-injury 
Measurement 
Timepoints 
Timepoint 1:  
1-11 days  
mean ± SD 
(7.4d± 2.2) 

Timepoint 2:  
~4 months 
(130.9d±14.5) 

Neuroimaging 
software & 
metrics 
AFNI, FSL, 
SPM, 
MATLAB 
NODDI 
toolbox 

Metrics: ODI, 
Vic and Viso 

Significant metrics and 
localization of effects 

Decreased Viso in mTBI 
compared to controls in 
the right post-central 
gyrus, right superior 
parietal lobule, right 
precuneus and left 
superior parietal lobule. 
Increased Vic in mTBI 
compared to controls in 
the right temporal pole, 
right inferior temporal 
gyrus, ILF, right interior 
temporal/fusiform gyrus, 
left fusiform gyrus, left 
middle occipital gyrus and 
bilateral precuneus.  
No differences observed 
for ODI.  
Use of a biologically 
informed microstructure 
diffusion toolbox 
algorithim  yielded 
different statistical 
relationships compared to 
standard NODDI 
analysis. 

 60 

 
 
 
 
 
Significant metrics and 
localization of effects 

Decreased NDI in self-
report poor sleeping 
mTBI patients compared 
to other groups in the 
first week following injury 
in the anterior TR, arcuate 
fasciculus, cing, CC, 
corticospinal tract, fronto-
pontine tract, IFF, ILF, 
optic radiation, parieto-
occipital pontine, 
striatofronto-orbital, 
striato-premotor, SLF, 
thalamic-parietal, 
thalamo-premotor tracts 
and UF.  
Decreased NDI in poor 
sleeping mTBI patients 
compared to other groups 
including good sleeping 
mTBI in the cing, optic 
radiation, striato-fronto-
orbital tract and SLF were 
associated with more 
symptoms on the post-
concussion symptom 
scale.  
Increased ODI in mTBI 
compared to controls in 
the left uncinate 
fasciculus.  
Decreased ISO in mTBI 
compared to controls in 
the CC, bilateral SLF 
direct and indirect 
segments.  

Table 2.3 (cont’d)  
Article  

Lima 
Santos et 
al., 2022 

Comparison 
group 
demographics 
Adolescent 
population 
sample mTBI:  
n= 57 (23F, 34M) 
Age: mean ± SD 
(15.3y ± 1.5)  

Controls:  
n = 33 (15F, 
18M) 
Age: (15.3y ± 1.6) 

Post-injury 
Measurement 
Timepoints 
Timepoint 1:  
Post-injury scan  
mean ± SD 
(6.9d± 2.5) 

Neuroimaging 
software & 
metrics 
FSL, and 
MATLAB 
NODDI 
toolbox, 
TractSeg for 
ROI 
generation 

Metrics: NDI, 
ODI, FISO 

Oehr et 
al., 2021 

Population 
sample mTBI:  
n= 26 (2F, 22M) 
Age: mean ± SD 
(34.81 ± 13.76)  

Tramua Controls:  
n = 20 (2F, 18M) 
Age: (38.75 ± 
12.59) 

Timepoint 1:  
6-12wk post 
injury:  

mTBI:  mean ± 
SD  
(73.31d ± 32.68) 

MRtrix3, FSL 
Comparison of 
WM tract 
averaged 
metrics with 
hand-drawn 
ROIs  

Control: (49.95d 
± 9.19) 

Metrics: ODI, 
ICVF, ISO 

 61 

 
 
 
 
 
 
Table 2.3 (cont’d)  
Article  

Shukla et 
al., 2021 

Muller et 
al., 2021 

Kawata et 
al, 2020 

Comparison 
group 
demographics 
Pediatric 
population 
sample mTBI:  
n= 320 (121F, 
199M) 
Age: mean ± SD 
(12.38 ± 2.42)  

Orthopedic 
Injury Controls:   
n = 176 (80F, 
96M) 
Age: (12.49 ± 
2.22) 
Population 
sample mTBI:  
n = 40 (28F, 
12M) 
Age: mean ± SD 
(48 ± 16.8) 

Controls:   
n = 17 (7F, 10M) 
Age: 33.2 ± 10.9 
Football Players: 
n = 17 (0F, 17M) 
Age: mean, range 
(16, 16-17) 

Significant metrics and 
localization of effects 

No differences observed 
in any diffusion metrics.  

Post-injury 
Measurement 
Timepoints 
Timepoint 1: 
mean ± SD 
(11.56d± 5.43) 

Neuroimaging 
software & 
metrics 
MRIcron, 
ExploreDTI. 
MATLAB 
NODDI 
toolbox  

Metrics: NDI, 
ODI, FISO 

Timepoint 1:  
mTBI: mean ± 
SD 
(73mo± 117.8) 

NODDI 
MATLAB 
toolbox, FSL 
TBSS-based 
voxel-wise 
analysis 

Metrics: ODI, 
Vic 

Decreased Vic in mTBI 
compared to controls in 
the bilateral SLF, ILF, 
IFF, and forceps major 
and minor.  
Decreased ODI in the 
bilateral SLF, ILF, IFF, 
and forceps major. 

Timepoint 1:  
Preseason 
baseline 
assessment w/ 
prior history of 
football season in 
last 12 mo 

MRtrix3, FSL 
TBSS-based 
voxel-wise 
analysis, 
NODDI 
MATLAB 
toolbox 

Serum Tau was negatively 
associated with NDI in 
the genu of the corpus 
callosum.  
Neurofilament light was 
negatively associated with 
NDI in the superior 
longitudinal fasciculus. 

Metrics: NDI, 
ODI 

 62 

 
 
 
 
 
Table 2.3 (cont’d)  
Article  

Palacios et 
al., 2020 

Wu et al., 
2018 

Comparison 
group 
demographics 
Population 
sample mTBI:  
n= 40 (9F , 31M) 
Age: mean ± SD  
(30.35 ±  7.5) 

Replication 
sample mTBI:  
n= 40 (11F , 
29M) 
Age: (34.38 ± 11)  

Orthopedic 
Controls:  
n = 14 (6F, 8M) 
Age: (31.71 ± 
10.14) 

Friend Controls: 
n = 19 (6F, 13 M) 
Age: (36.33 ± 
13.5) 
Patients:  
mTBI:  
n= 19 (11F , 8M) 
Age: mean ± SD 
(35 ± 12) 

Controls:  n= 23 
(11F , 12M) 
Age: (35.6 ± 14.1) 

Post-injury 
Measurement 
Timepoints 
Timepoint 1: 
2 wk mean ± SD  
(13.30d± 2.10) 

Timepoint 2:  
6 mo 
(184d± 8.86) 

Neuroimaging 
software & 
metrics 
NODDI 
MATLAB 
toolbox, FSL 
TBSS-based 
voxel-wise 
analysis  
JHU-atlas 
based ROI 
analysis 

Metrics: ODI, 
NDI and 
FISO 

Timepoint 1: 
mTBI: mean ± 
SD (15d ± 10) 

Control: (31d ± 
20) 

TBSS-based 
voxel-wise 
analysis  
JHU-atlas 
based ROI 
analysis  
Metrics: Vic, 
OD, and P0 

Significant metrics and 
localization of effects 

Initial mTBI cohort: 
Increased FISO and 
decreased NDI at 2-weeks 
in widespread brain 
regions in mTBI 
compared to controls.  
Decreased NDI 
longitudinally from 2-
week to 6-month 
timepoint in the anterior 
corona radiata, posterior 
corona radiata, posterior 
TR, ILF, and IFF, 
anterior TR, EC, and UF.  
Increased FISO was 
associated with 
membership to a cluster 
with greater improvement 
on a global improvement 
measurement in an 
unsupervised machine 
learning analysis. 

Decreased Vic  in mTBI 
compared to controls in 
the CC, anterior and 
superior CR.  
Decreased Vic and P0 were 
associated with worse 
performance on 
neuropsychological tests.  
Increased Vic and P0 were 
associated with better 
performance in the 
trauma control group. 

 63 

 
 
 
 
 
 
  
 
 
Table 2.3 (cont’d)  
Article  

Churchill 
et al., 2019 

Churchill 
et al., 2017 

Comparison 
group 
demographics 
Patients:  
mTBI:  
n= 33 (17F, 16M) 
Age: mean ± SD 
(20.5 ± 1.7) 

Controls:  n = 33 
(17F, 16M) 
Age: (20.3 ± 2.0) 

Athletes with 
prior history of 
concussion:  
n = 31 (16F, 
15M) 
Age: mean ± SD 
(21 ± 1.7) 

Athletes without 
prior history of 
concussion:  
n = 37 (20F, 
17M) 
Age: (20.1 ± 1.7) 

Post-injury 
Measurement 
Timepoints 
Timepoint 1: 
symptomatic 
phase 1-7 days 
post injury 

Timepoint 2:  
following medical 
clearance for 
return to play 
mean, IQR (19d, 
13d-58) 

Neuroimaging 
software & 
metrics 
TBSS-based 
voxel-wise 
comparison, 
in-house N-
way partial 
least squares 
testing of the 
six diffusion 
metrics.   
Metrics: FA, 
AD, RD, ODI, 
VISO, VIC 

Timepoint 1:  
Preseason 
assessments  

FSL, NODDI 
MATLAB 
Toolbox 

Metrics: VIC, 
ODI 

Significant metrics and 
localization of effects 

Decreased VIC in mTBI 
compared to controls in 
the CR and longitudinal 
fasciculus at the first 
timepoint.  
These differences 
remained significant at the 
second timepoint, but 
were not significant when 
comparing longitudinal 
change between 
timepoints.  
Increased ODI between 
the first and second 
timepoint in the UF.  
Decreased ODI in 
athletes with history of 
concussion compared to 
those without in the genu 
body and splenium of CC, 
Fx, anterior and posterior 
limb of IC, left CR.  

Increased VIC in athletes 
with history of 
concussion compared to 
those without in the genu 
body and splenium of CC, 
Fx, anterior and posterior 
limb of IC, left CR. 

VIC increased and ODI 
decreased with increasing 
time since last concussion 
between subjects with a 
history of concussion.  

 64 

 
 
 
 
 
 
 
 
 
Table 2.3 (cont’d)  
Article  

Mayer et 
al., 2017 

Comparison 
group 
demographics 
Mixed Martial 
Arts professional 
athletes  
rmTBI:  
n = 13 
(11M, 2F) 
Age: mean ± SD 
(28.2± 4.9) 

Sex, age and 
education 
matched 
Controls: n = 14 
(12M, 2F) 
Age: (28.1± 5.1) 

Post-injury 
Measurement 
Timepoints 
Timepoint 1: 
Scanned during a 
competition 
season 

Neuroimaging 
software & 
metrics 
SPM, FSL, 
Voxel-wise 
comparison 
with AFNI 
Metrics: VISO, 
VIC and ODI  

Significant metrics and 
localization of effects 

Increased ODI in rmTBI 
compared to controls in 
grey matter regions 
including the caudate and 
cerebellum, and WM 
regions including the 
splenium of the CC, cing 
and posterior CR.  
Decreased VISO in rmTBI 
compared to controls in 
grey matter regions 
including the 
diencephalon, and WM 
regions including the 
brain stem and cerebellar 
peduncles. 
Increased VISO in rmTBI 
compared to controls in 
WM regions including the 
IC, EC, CR, saggital 
striatum, posterior TR 
and SLF.  
Increased VIC in the 
cerebellar peduncles, IC 
and SLF.  

Constrained Spherical Deconvolution and Fixel-Based Analysis in mTBI 

Early studies that utilized CSD mainly took advantage of the improved tractography 

results produced by tracking with CSD-estimated FODs.171 For example, CSD was used to analyze 

the TBICare study, a prospective study of mTBI in Europe that collected a single timepoint in the 

subacute period following injury.172 This early study highlights the utility of CSD as a method for 

improved tractography and improved sensitivity and specificity by accounting for voxels with 

crossing fibers. In this analysis, DTI metrics in an FA WM skeleton generated using TBSS were 

compared between groups, in addition to a skeleton with only voxels that contained a single fiber 

population as determined by CSD. The global average of metrics in the whole brain tractogram 

 65 

 
 
 
generated with probabilistic tractography was also compared. Interestingly, using CSD, it was 

determined that only 29.13% of voxels in the FA WM skeleton contained a single fiber population. 

This would suggest that even methods like TBSS aimed at reducing partial volume effects are still 

likely to contain significant confounds from crossing fibers. The mean FA in voxels of the whole 

WM skeleton was 0.412 in mTBI patients vs. 0.576 in voxels containing only a single fiber 

population, demonstrating the impact of crossing fibers on summary measures like FA.  

Another study employed CSD-based whole-brain tractography to conduct graph theory 

analysis in a sample of mTBI patients based on the number of post-traumatic symptoms in the 

subacute stage of recovery.173 Whole brain structural connectivity was similar between mTBI 

subjects and controls. However, mTBI subjects had a trend relationship between lower network 

clustering and higher processing speed, which was associated with reporting >3 symptoms post-

injury and at least one cognitive and/or affective complaint. The improved performance of 

tractography algorithms on DWI signal modeling with CSD may facilitate the use of more 

sophisticated graph analysis techniques in future studies. More recently, a study compared CSD-

based tractography with the automated probabilistic segmentation tool TractSeg with DTI-based 

deterministic tractography methods conducted by board-certified neuroradiologists. They found that 

when comparing tract-averaged FA values in tracts generated with each method, there were 

significantly lower FA values in the mTBI group compared to controls for tract averages created 

using TractSeg-generated FA values. No significant differences were observed for DTI-based tract 

averages. They also conducted this study on low b-value data, collecting only 15 gradient encoding 

directions at b= 800 mm/s2, a more clinically feasible acquisition.   

The first study to employ a FBA in mTBI patients was conducted by Wallace and colleagues 

in 2020. This study examined WM changes at roughly 7 months post-injury.174 No differences 

between mTBI patients and controls were identified. A possible explanation for this result is the 

 66 

 
 
highly variable scanning time and length of scanning time employed in this study (Table 4). 

Additionally, the majority (63%) of the mTBI sample had a Glasgow coma scale score of 15, 

meaning this mTBI sample may have sustained very minor injuries.  

Two more recent analyses of sports-related mTBI in samples of Australian football players 

found significant differences in fixel-based metrics in widespread brain regions. The first study by 

Wright and colleagues compared football athletes with mTBI and athlete controls in addition to 

conducting within-group sex comparisons.175 They observed a pattern of increased FD that 

improved between the first timepoint within 24-48 hours of injury and a second timepoint at 2 

weeks post-injury. The whole brain fixel-wise analysis revealed male athletes had increased FD in the 

cingulum compared to female athletes sustaining a mTBI. The cause of sex differences needs further 

exploration to determine if morphological differences between sexes or some combination of 

factors lead to increased susceptibility of one sex over another. DTI metrics were also compared in 

this study; interestingly, differences were detected in more widespread regions than with the FBA 

metrics.  

A second study compared subgroups of mTBI patients scanned ≤ 12 days following injury, 

>12 days, and controls. Interestingly, acute mTBI patients had increased FD, FC, and FDC 

compared to subacute mTBI patients.176 The authors posit this could be related to cytotoxic edema 

following injury in the acute period, resulting in axonal swelling, which resolves in the subacute 

period. There was also a preponderance of left-sided changes observed, which may reflect 

differential hemispheric susceptibility via a mechanism that is currently unclear. The findings of this 

study also seem to suggest a decrease in FBA metrics that occurred in the subacute period, which 

could indicate either recovery following injury or axonal loss following injury. Future studies will 

benefit from larger sample sizes, more frequent and standard data collection following injury, and 

 67 

complimentary signal modeling techniques to build a complete picture of WM changes following 

mTBI.  

Table 2.4 Summary of Studies Utilizing FBA and/or CSD in mTBI Patients 

Author 

Sample 
Characteristics 

Tallus et al., 
2023 

Population 
sample  
mTBI:  
n = 37 (15F, 
22M) 
Age: mean ± SD 
(37.2 ± 11.4) 

Age and Sex 
matched 
Controls: n= 41 
(17F, 24M) 
Age: (36.4 ± 
11.9)  

Post-injury 
Measurement 
Time 
Timepoint 1:  
Mean, range 
(1.2y, 2.2wk – 
9.9y) 

DWI Image 
Analysis Method 

Location of Significant 
Effects 

MRTrix3 
standard 
preprocessing. 
FOD peaks for 
Tract-Seg 
analysis. 

Metrics: FA, 
MD 

Fixel-based analysis 
metrics not calculated 
in this manuscript.  
CSD probabilistic 
tractography was able 
to differentiate mTBI 
from controls using 
tract-wise averages of 
FA, while DTI 
deterministic 
tractography was not.  

 68 

 
 
 
 
 
Post-injury 
Measurement 
Time 
Timepoint 1: 
mTBI scanned 
≤ 12 days 
mean ± SD 
(7.6d ± 3.3) 

Timepoint 2: 
mTBI scanned 
>12 days: 
(34.3d ± 19.4) 

Table 2.4 (cont’d)  
Author 

Sample 
Characteristics 

Mito et al., 
2022 

Patients:  
mTBI scanned ≤ 
12 days: n = 14 
(0F, 14M) 
Age: mean ± SD 
(24.07 ± 3.82) 

mTBI scanned 
>12 days: n=15 
(0F, 15M) 
Age: (25.18 ± 
3.45) 

Age matched 
Controls: n=29 
(0F, 29M) 
Age: (24.1 ± 4.8) 

DWI Image 
Analysis Method 

Location of Significant 
Effects 

MRTrix3 
standard 
preprocessing 
and fixel-based 
analysis. 
Dhollander and 
multi-shell 
multi-tissue 
algorithms used. 

Metrics: FD, FC 
and FDC 

Increased FD in the 
acute group (scanned ≤ 
12 days post injury) 
compared to the 
subacute (>12 days) 
and control groups in 
the left posterior 
parahippocampal WM 
extending into the 
isthmus.  
Increased FD in acute 
compared to subacute 
specifically in the 
splenium of the CC, 
and left posterior 
parahippocampal WM. 
Increased FC in the 
acute group compared 
to subacute group in 
the splenium and left 
frontal aspect of the 
SLF.  
Increased FDC was 
observed in the acute 
group compared to the 
subacute and control 
groups predominantly 
in the CC and 
prominently in the 
forceps minor.  

 69 

 
 
 
 
 
Table 2.4 (cont’d)  
Author 

Sample 
Characteristics 

Roine et al., 
2022 

Population 
sample  
mTBI:  
n = 85 (26F, 
59M) 
Age: mean ± SD 
(47 ± 20) 

Age and Sex 
matched 
Controls: n= 41 
(22F, 18M) 
Age: (50 ± 20) 

Post-injury 
Measurement 
Time 
Timepoint 1:  
3wks post-
injury 
mean ± SD 
(21.14 ± 
14.91) 

Timepoint 2: 
8mos 
(251.8 ± 
86.87) 

Wright et al., 
2021 

Patients:  
mTBI n=14 (6F, 
8M)  
Male Age: mean 
± SD (22.1 ± 2.4) 
Female Age: 
(24.2 ± 4.1) 

Timepoint 1: 
24-48h post 
injury 

Timepoint 2: 
2 weeks post 
injury   

Athlete healthy 
controls: n= 16 
(7F, 9M) 
Male Age: (24.7 ± 
2.4) 
Female Age: 
(24.3 ± 4.3) 

DWI Image 
Analysis Method 

Location of Significant 
Effects 

MRtrix3, FSL, 
FreeSurfer. 
MATLAB 
connectivity 
toolbox 

Metrics: Average 
betweenness 
centrality, 
Normalized 
clustering 
coefficient, 
normalized 
global efficiency, 
Normalized 
characteristic 
path length, 
Small-worldness, 
Average degree, 
Average strength 

MRTrix3 
standard 
preprocessing 
and fixel-based 
analysis. Tract 
based ROI 
analysis. 
Dhollander and 
MSMT 
algorithms used.  

Metrics: FD, FC 
and FDC 

No differences in 
global network 
properties between 
patients with mTBI 
and control subjects at 
the either timepoint. 
Increased betweenness 
centrality in mTBI 
compared to controls 
in the R pars 
opercularis at the 
second timepoint. This 
did not remain 
significant when 
comparing CT-negative 
patients and controls.  
Decreased normalized 
global efficiency and 
increased small 
worldness in mTBI 
patients with GOSE of 
7 or 8 when compared 
to GOSE <7. 
Increased FD in mTBI 
compared to controls 
in the splenium of the 
CC with number of 
significant fixels 
observed decreasing 
from the 24-48h to the 
2 week timepoint.   
No significant 
differences observed 
for FC or FDC at 
either timepoint when 
comparing mTBI and 
controls.  
Increased FD in the 
cing of male compared 
to female mTBI 
patients at both 
timepoints. 

 70 

 
 
 
 
 
 
Post-injury 
Measurement 
Time 
Timepoint 1: 
mean ± SD 
(194.4d± 38.3) 

DWI Image 
Analysis Method 

Location of Significant 
Effects 

No significant 
differences in FD, FC 
and FDC when 
comparing mTBI and 
controls 

MRTrix3 
standard 
preprocessing 
and fixel-based 
analysis. 
Tournier and 
CSD algorithms 
used 

Metrics: FD, FC 
and FDC 

Timepoint 1: 
3+ 
Posttraumatic 
symptoms 
mTBI: 
Median, range 
(32d, 22-56) 

≤ 2  
Posttraumatic 
symptoms 
mTBI: (33d, 
22-69) 

Graph Theory 
Analysis with 
Explore DTI 
and MATLAB 
brain 
connectivity 
toolbox.  

Metrics: Eglob, 
Ci, C, g, Eloci, 
Eloc, Q, Ki, BCi. 
ECi 

Fixel-based analysis 
metrics not calculated 
in this manuscript.  
Lower eigenvector 
centrality within the 
left temporal pole in 
mTBI compared to 
controls.  
Whole brain structural 
connectivity was 
similar for controls and 
mTBI 

Table 2.4 (cont’d)  
Author 

Sample 
Characteristics 

Wallace et al., 
2020 

van der Horn 
et al., 2017 

Patients:  
mTBI: n=133 
(30F, 103M) 
Age: mean ± SD 
(43.6 ± 16.9) 

Overall Controls:  
n=107 (47F, 
60M) 
Age: (39.3 ± 
16.6) 

Healthy Controls 
n=60  

Orthopedic 
Controls: n=47 

Patients:  
3+ Posttraumatic 
symptoms mTBI: 
n= 33 (16F, 17M) 
Age: median, 
range (33, 19-63)  

≤ 2 
Posttraumatic 
symptoms mTBI: 
n=34 (3F , 30M) 
Age: (34, 20-64) 

Age, Sex and 
Education 
matched 
Controls:  n= 20 
(6F, 14M) 
Age: (30, 18-61) 

 71 

 
 
 
 
 
 
 
 
 
Post-injury 
Measurement 
Time 
Timepoint 1: 
mTBI:  mean 
± SD (21d± 
15) 

Table 2.4 (cont’d)  
Author 

Sample 
Characteristics 

Mohammadian 
et al., 2017 

Patients:  
mTBI: n= 102 
(32F, 70M) 
Age: mean ± SD 
(47 ± 20) 

Orthopedic 
Controls:  n= 30 
(16F, 14M) 
Age: (50 ± 20) 

DWI Image 
Analysis Method 

Location of Significant 
Effects 

Fixel-based analysis 
metrics not calculated 
in this manuscript.  
ICC values 
demonstrated limiting 
comparisons to single 
fiber population voxels 
in the WM skeleton 
was superior to 
comparisons in the 
whole brain tractogram 
and FA WM skeleton 

Bespoke analysis 
with MATLAB 
and 
ExploreDTI. 
Voxel-wise 
comparison of 
TBSS FA 
skeleton, voxels 
with a single 
fiber population 
as determined 
with CSD 
CSD tractogram 
based ROI 
analysis 

Metrics: FA, 
MD, AD, RD 

General Discussion 

Taken together, these DWI studies of mTBI with various modelling techniques demonstrate 

WM differences compared to healthy controls throughout the recovery period following mTBI. 

Interpreting these changes is complex, as observed differences could be attributed to various 

phenomena at the microstructural level and to numerous sources of variance, including post-injury 

measurement time, acquisition parameters, scanner differences, and statistical methods. Nonetheless, 

many studies report differences in DKI and NODDI while observing no differences in DTI 

metrics, demonstrating the increased sensitivity of these measures in mTBI. Many studies had 

relatively small sample sizes and differing methodologies, highlighting that the field would benefit 

from more consistency in analysis techniques and study design. Still, an early look at this developing 

body of literature that utilizes these contemporary DWI analysis techniques has revealed promising 

and exciting findings.  

 72 

 
 
 
DKI studies focused on athletes with sports-related mTBI were, on average, younger than 

the general population. Sports-related mTBI and mTBI from other etiologies are two areas of 

literature that can inform one another and may unlock new questions as more details about WM 

change from these two distinct populations and injury mechanisms emerge. This review noted that 

uniformly, studies in sports-related mTBI observe an acute decrease in the DKI metrics MK, RK , 

and AK, while the opposite is observed in population-based samples. A decrease in MK results from 

a more uniform environment of diffusion, which could be related to degenerative changes in WM 

and axonal shrinkage, whereas the increase could be due to tighter packing of fiber bundles due to 

edema. Results from several studies suggest changes in DWI metrics in sport-related mTBI may 

relate to accumulated damage or altered homeostatic WM changes from cumulative contact-sport 

participation as opposed to the single traumatic event resulting in diagnosis with mTBI.154–156  

A major critique of DKI is that despite its increased sensitivity to underlying tissue 

microstructural complexity, it still has limitations in interpretability. It is also susceptible to partial 

volume effects, which limits its reliability in whole-brain VBA, especially at tissue interfaces like the 

grey-white matter junction. Techniques developed for DTI to address similar limitations, such as 

TBSS, may further obscure findings and complicate interpretation. By design, TBSS only questions 

part of the data collected in WM, resulting in loss of information, computing artificially lower FA 

values, and making it difficult to colocalize effects in multimodal studies.105,138,172,177 Future work will 

benefit from reporting multiple diffusion metrics and employing multi-level statistical comparison 

schemes, as shown in the work of Goubran et al., 2023 and Huang et al., 2022. Goubran and 

colleagues’ work on male football players showed ODI, but not DKI metrics was sensitive to player 

position. This work similarly highlights the importance of repeated sampling within subjects, as 

dynamic temporal WM changes even in adulthood may obscure true patterns of pathologic change 

in experimental comparison groups.178 

 73 

NODDI’s increased sensitivity and specificity come at the cost of the highly computationally 

intensive analysis process, with reported computation times greater than 65 hours without using a 

high-powered computing cluster.179 This is partly due to its dependence on multi-shell data with 

many diffusion encoding directions. An ongoing body of work is actively attempting to create 

analysis schemes that would allow the NODDI model to fit low b-value or single shell 

acquisitions.180 Additionally, axon tracing in animal studies has brought into question the degree to 

which WM regions have a coherent morphology that would facilitate Gaussian modeling in any 

significant part of the brain. In animal work with electron microscopy, the corpus callosum, a region 

commonly thought to have a uniform orientation of axons, had significant dispersion in a mouse 

brain.181 Ground truth work in human brains to better understand the extent of crossing fibers is still 

needed and will inform future modeling techniques.182,183 Despite the increased complexity of this 

biophysical model, the extracellular compartment is still modeled using a Gaussian distribution, 

which may only be an accurate representation in small regions of the brain’s WM. Finally, as 

discussed previously, the use of fixed intra- and extra-axonal diffusivity values may bias this model in 

certain regions. However, novel optimization methods have been developed for a more biologically 

accurate representation.165 

Few studies have been published using CSD, and only 3 have conducted FBA in mTBI 

patients. This may be partly due to it being a relatively recently released software package, or the 

computational skill to implement this method.18 A critique of CSD is the assumption of a single fiber 

response, which, similarly to the above, means this method is only an approximation of the brain’s 

WM.105 Additionally, to be clinically useful, large standard population-based studies would need to 

be conducted to establish population average tissue response functions for normative comparisons 

of single patients. Nonetheless, these early studies seem to support an acute pattern of edema 

 74 

 
resulting in increases in FD and FDC, followed by axonal degeneration in the subacute period as 

indicated by decreases in FD, FC and FDC. These changes need to be confirmed in future studies.176  

Recent animal studies provide strong evidence that the interpretation of microstructural 

change in human studies utilizing biophysical models is accurate. For example, a recent study 

collected 7T DWI data of ex vivo fixed brain tissue in a mouse model of closed head injury.184  This 

study observed increased ODI in mTBI animals in the optic tract only, compared to sham, with no 

significant differences observed for FA.184 Interestingly, behavioral impairment resolved after a 6-

week timepoint, whereas alterations in ODI and pathologic staining with GFAP persisted to an 18-

week timepoint post-injury. Similar results were observed in a smaller animal study of the acute 

period following injury, in which increased NDI and ODI compared to control animals were 

observed at 1- to 4-hour post-injury timepoints. Similarly, this study observed no differences in any 

DTI metrics.185  

The strongest evidence from animal studies has been work by Chary and colleagues.111 In 

this study, colocalized DWI maps of DTI, NODDI, and FBA metrics at 11.7T and 

photomicrographs of nissl and gold chloride stained histologic sections were compared.111 In this ex 

vivo study, more significant voxels were observed between mTBI and sham mice with decreased FD 

and FDC when compared to FA in the WM. Most importantly, no differences were observed for 

FWF or NDI in this data collected in the subacute stage 35 days post-injury. In fact, the only 

significant NODDI metric in this animal study was ODI, which was observed in regions near the 

fluid percussion injury. This may indicate that for longitudinal studies, FBA metrics are more 

sensitive than NODDI metrics in identifying changes in the subacute stage, whereas the FWF of 

NODDI is more useful in the acute period following injury. 

Finally, a clear observation in this review is the underrepresentation of females. Among the 

reviewed manuscripts, 11 samples included no females, and of the 28 studies that did include 

 75 

 
 
 
 
 
 
females, 15 had a greater proportion of males than females. This pattern of underrepresentation is in 

line with the recent consensus of a national work group and represents an additional emerging gap 

in our understanding of sex differences in WM change following mTBI.186 The work by Wright and 

colleagues suggests that in a sample of biologically female Australian football athletes, less detectable 

WM changes were observed compared to males.175 Larger comprehensive studies like TRACK-TBI 

have concluded females experience more symptoms, including insomnia, with greater severity and 

are more susceptible to persistent symptoms following mTBI.187,188 Taken together, there may be a 

lower threshold of damage for symptom generation in females and more pronounced differences in 

adult females as compared to young females that need further investigation. Many of the included 

studies utilized age, sex and education matched controls, which seems to be an ideal practice, and 

even large studies have shown success collecting friend controls (Tables 4.2-4.4).112,142–145,151–153,159,189–

191    

Conclusion 

The overall results of these studies suggest there may be unique trajectories of WM recovery 

that have yet to be fully described by Tensor based methods. The improved sensitivity and 

specificity of higher-order modeling methods, paired with the widespread adoption of high angular 

resolution data collection, will help to delineate these individual differences in recovery trajectory. 

The increased understanding of WM changes in healthy individuals will also improve our 

understanding of neurotrauma. FBA is unique in that it offers a novel volume element for the 

analysis of diffusion data, which may help to further our understanding of differential effects on 

non-dominant fiber tracts in a voxel and how they are affected by trauma. Similarly, NODDI moves 

away from modeling the diffusion signal as a single Gaussian compartment and adds increased detail 

that brings this representation closer to biological reality. DKI is clearly a distinct improvement over 

the original DTI model. However, the growing body of literature suggests that, like DTI, DKI is still 

 76 

unable to provide the clear picture of WM change following mTBI that is necessary to translate the 

modality into clinical use. Future work will benefit from utilizing a combination of several models 

and analysis techniques to characterize WM change in mTBI. Additionally, legacy datasets could be 

revisited and analyzed with contemporary techniques if made available via data-sharing initiatives 

such as the Federal Interagency Traumatic Brain Injury Research (FITBIR) Informatics System.192 

 77 

 
  
 
 
 
 
 
 
 
CHAPTER 3: FIXEL-BASED ANALYSIS PIPELINE TESTING IN A TRACK-TBI 
SUBSAMPLE 

INTRODUCTION 

 In the present study, a subsample of the TRACK-TBI comprehensive assessment plus MRI 

(CA+MRI) group was randomly selected from a single recruitment site as an initial test of the 

standard FBA pipeline preprocessing and analysis steps. Diffusion MRI data that was collected on 

the same scanner was analyzed with the standard FBA pipeline to determine the feasibility of its use 

with concatenated data from two shells. The original acquisition recommended for single tissue 

constrained spherical deconvolution (CSD) was a single high b-value of 3000 mm/s2 or greater; 

however, the ideal data acquisition protocol is debated.193–195 The present study is a secondary 

analysis of participants with two separate acquisitions of b=1300 mm/s2 and b=3000 mm/s2  

obtained in the same scanning session, which, if concatenated, could represent a pseudo-multi-shell 

acquisition. Recently, a multi-shell, multi-tissue (MSMT) CSD pipeline was proposed that allows 

deconvolution with three separate tissue response functions for white matter (WM), grey matter 

(GM), and cerebrospinal fluid (CSF).193 This model has shown good reliability across multiple 

timescales with varying acquisitions, including an interclass correlation coefficient larger than 0.8 for 

all tissue compartments for data collected >100 days apart.196 Therefore, we sought to apply the 

MSMT-CSD pipeline to the pseudo-multi-shell TRACK-TBI data and apply the recently described 

longitudinal FBA approach to the 2-week and 6-month time points.   

Few studies have used CSD and FBA to analyze single-shell DWI data in mTBI.79,176,214 In 

addition, to our knowledge, no studies have applied MSMT-CSD to multi-shell DWI data in mTBI 

patients and analyzed it following the standard FBA pipeline outlined by the developers of MRtrix3. 

In the present study, we seek to apply the standard pipeline for MSMT-CSD and FBA to DWI data 

collected at a single site on a single scanner during the TRACK-TBI study. We also sought to 

 78 

 
 
 
 
examine change in WM tracts using a longitudinal FBA approach recently outlined in the 

literature.197 Finally, we apply a novel deep learning algorithm, Synthesized b0 for diffusion 

distortion correction (Synb0-DisCo), which facilitates susceptibility field distortion of diffusion MRI 

data that was collected without reverse phase-encoded images, as is the case in the legacy TRACK-

TBI data.113,198 Data used for the present study are available publicly through the FITBIR registry 

(https://fitbir.nih.gov/) from the TRACK-TBI U01 Study [FITBIR-DATA0012173].192  

MATERIALS & METHODS 

Transforming Research and Clinical Knowledge in Traumatic Brain Injury Overview 

The TRACK-TBI Study is a multi-site clinical cohort study that collected patient data across 

the spectrum of TBI, including mild, moderate, and severe.15,112 The present study will focus on the 

neuroimaging data collected from participants in the CA+MRI sub-cohort, which includes 600 

participants.112 The CA+MRI group had anatomical MRI and DWI collected at 2-weeks and 6-

months post-injury.113  

TRACK-TBI: Participants 

Participants diagnosed with a TBI based on the American College of Rehabilitation 

Medicine1 <24 hours before presenting at an emergency department were included if they met the 

following criteria: received a brain MRI/CT, had adequate visual acuity and hearing for testing, and 

were fluent in English or Spanish in order to provide consent. Patients were excluded if they met 

one or more of the following criteria: polytrauma present, penetrating injury to the skull, spinal cord 

injury with American Spinal Injury criteria C or worse, low likelihood of follow-up, an existing 

debilitating mental health or neurologic disorder, notable preexisting health conditions, 

contraindications to MRI or were incarcerated, pregnant, awaiting psychiatric evaluation.112 For the 

present study, patients were considered to have a mTBI if their Glasgow coma scale (GCS) was 13-

15 in the CA+MRI group (N= 554). Of the 554 who met the mTBI GCS metric, 86 had two-shells 

 79 

 
of diffusion-weighted magnetic resonance imaging data collected (N= 86). We used all available 

control multi-shell data which was only collected at the Zuckerberg San Francisco General Hospital 

and Trauma Center recruitment site for cases and controls. 20 mTBI patients with available multi-

shell data were randomly selected for pipeline testing. Image quality and model fit was assessed at 

intervening steps throughout the analysis pipeline to determine successful application of the 

preprocessing and analysis steps.   

TRACK-TBI: Neuroimaging Acquisition  

Neuroimaging data were collected at 2-weeks and 6-months post-injury at the same clinical site 

on the same 3T MRI scanner. Whole-brain high-definition fiber tracking DWI data was collected (64 

directions at b = 1300 and 3000 s/mm2), eight acquisitions at b = 0 s/mm2 for each set of 64 

directions. 

Neuroimaging Preprocessing and Analysis 

The imaging preprocessing and analysis pipeline utilized in this pilot analysis is summarized 

in Figure 3.1 adapted from a figure in Genc et al., 2018, to signify our modified preprocessing. All 

developer recommendation for default parameters were followed, and detailed command-line 

analysis step descriptions are available online at 

(https://mrtrix.readthedocs.io/en/latest/fixel_based_analysis/mt_fibre_density_cross-section.html,  

Accessed Nov 2022). The standard preprocessing for MSMT-CSD included denoising, Gibb’s 

ringing artifact removal, eddy current, and susceptibility field distortion correction. Since reverse-

phase encoded images were not collected during the initial study, Synb0-DisCo183 was utilized to 

generate synthetic undistorted b0 images for input to FSL’s topup option in eddy. Finally, individual 

subject images were corrected for bias field distortion to improve mask estimation directly from the 

DWI data with dwi2mask. 

 80 

 
Tissue response functions for WM, GM and CSF were estimated from preprocessed DWI 

data with dwi2response using the Dhollander algorithm option.199,200 With the voxels option of 

dwi2response, the voxel.mif file was overlayed on the DWI data and visually inspected to ensure the 

automatically selected voxels were true single tissue voxels. All subject tissue response functions 

were then averaged for the whole sample to create average tissue response functions for the WM, 

GM, and CSF. DWI images were then upsampled with mrgrid to the recommended isotropic voxel 

size of 1.25 mm3. Brain masks were estimated directly from the subject DWI data using dwi2mask. 

Fiber orientation distribution (FODs) functions were estimated with dwi2fod using the single set of 

averaged response functions with the MSMT-CSD algorithm (msmt_csd) option.193 A colorized 

image based upon the three tissue types was then generated, and subject FODs were overlayed on 

this image to ensure WM fixels were only fit in WM regions and that regions containing crossing 

fibers had multiple fixels fit per voxel (Figure 3.1). A second bias field correction and intensity 

normalization was conducted on the FODs to approximate signal amplitude across subjects and 

correct for sample-wide intensity differences not corrected in previous individual bias field 

correction preprocessing steps. 

 81 

 
Figure 3.1- Schematic representation of standard preprocessing and analysis pipeline for fixel-based 
analysis adapted from Genc et al., 2018. This schematic includes the modified preprocessing for 
images without reverse phase encoded b0 images collected with the use of Synb0-DISCO, and 
representative intermediate images throughout the processing pipeline. (Top Row Fourth Column)- 
axial view of preprocessed DWI b0 image after eddy current and motion correction, (Second Row 
First Column)- Average b0 image for a single subject colorized by voxels selected for three tissue 
response function estimation with white matter (WM) FODs overlayed for quality assessment of fit in 
region with crossing fibers, and minimal fitting in the CSF and grey matter, (Third Row Fourth 
column)- axial view of the whole sample WM FOD template generated from the linear and non-linear 
registration of all intra-timepoint subject templates, (bottom row third column)- coronal view of whole 
sample WM FOD template overlayed with tractography generated from this image. 

 The procedure described by Genc and colleagues, 2018 to create a whole sample unbiased 

template for longitudinal FBA was implemented.197 First, each individual subject's first and second 

timepoint FODs were brought into an intra-template space using the population_template script. These 

intra-subject templates were used as input a second time to the population_template script to generate a 

whole sample template unbiased towards either timepoint.201  

 82 

 
 
 
Subject warps into template space for all FOD images were calculated with mrregister.202 

Warps were applied to all subject masks with mrtransform and averaged to create a whole sample 

template mask. The whole sample mask in template space was visually inspected to ensure minimal 

extra-brain regions were included and no holes or fissures would exclude WM regions from 

subsequent analysis. A whole sample fixel mask was generated from the whole sample template 

FOD image, thereby defining the fixels that would be included in subsequent analysis steps with 

fod2fixel. Subject fixels were then segmented from their warped FOD images in the whole sample 

template space, and FD was calculated per fixel in each voxel with fod2fixel. Subject fixels were then 

reoriented in whole sample template space with fixelreorient and assigned to template fixels with 

fixelcorrespondence, specifying which fixels across subjects’ match fixels in template space. FC was 

computed from each subject's calculated warp to template space with warp2metric, and this metric 

was normalized by applying a logarithmic transformation to each image.203,204 The FDC measure was 

then calculated as the product of the fiber density and cross-section.  

Whole brain tractography was then conducted on the whole sample population template 

FOD image using tckgen with 20 million streamlines. Spherical-deconvolution informed filtering of 

tractograms was then performed on the 20 million streamline image with tcksift to generate a filtered 

tractogram with 2 million streamlines. This tractogram was used to generate a fixel-fixel connectivity 

matrix with fixelconnectivity. 

Statistical Analysis  

As this is a population-based sample and WM is known to change as adults age, demeaned age 

was included as a nuisance covariate in our statistical model. Sex was also included as a potential 

confound in statistical models. We tested the hypotheses that the mTBI subjects would have decreased 

FD, FC, and FDC at the 2-week and 6-month timepoint. We also calculated the difference between 

 83 

 
 
 
the second and first timepoints and divided by the amount of time in years between each participant’s 

two scans to get a change image representative of longitudinal change in FBA metric (Equation 3.1).  

∆𝐹𝐵𝐴	𝑀𝑒𝑡𝑟𝑖𝑐 = 	

𝑚𝑒𝑡𝑟𝑖𝑐)! − 	 𝑚𝑒𝑡𝑟𝑖𝑐)*
𝐷𝑎𝑡𝑒	1 − 𝐷𝑎𝑡𝑒	2	(𝑦)

(3.1) 

We tested the hypothesis that change in FBA metrics would be greater in mTBI patients compared to 

controls. Permutation testing with 5000 repetitions was utilized for the reported results of the FBA, 

and family-wise error-corrected p-values were computed for each comparison. 

RESULTS 

On the whole brain tractogram (Figure 3.2), differences in effect size (Cohen’s D) at 

(p<0.01) before family-wise error correction between controls and mTBI patients are displayed. 

Differences are highlighted in yellow, representing a positive effect in which the metric is higher in 

the mTBI group compared to the controls. Differences highlighted in blue represent the opposite 

effect where the metric is lower in the mTBI group compared to controls. Differences representing 

decreased FD, FC, and FDC are apparent peripherally in the cortical WM, primarily in the frontal 

WM regions. Most interestingly, there are large effects in deep WM, specifically in the anterior limb 

of the internal capsule, external capsule, and the forceps minor. No effects remained significant after 

family-wise error correction for FD, FC or FDC at either timepoint or in change images. 

Table 3.1 Summary of Findings in Chapter 3  
Main Finding 
1 

No significant differences between mTBI and control groups for any fixel based 
metrics. Uncorrected differences between mTBI and control groups revealed a 
typical pattern of regional differences with lower white matter microstructural 
metrics in the mTBI group in the frontal and deep white matter regions.  This is 
an expected pattern of injury due to the contact of the brain with the frontal parts 
of the skull, and the torsional forces of impact which are greatest in the regions 
closest to the axis of rotation about the brainstem.  
Synb0-DisCo was not successful in all participants and introduced geometric 
distortions, it was removed from the analysis pipeline in Chapter 4 after quality 
check of the raw images revealed only modest susceptibility field distortions were 
present in some participants.  

Main Finding 
2 

 84 

 
 
 
 
 
 
Figure 3.2.-  Results of MRtrix3 fixelcfestats command comparing mTBI patients and controls are displayed for the FBA metrics fiber 
density, fiber cross-section and fiber density cross section at the 2-week and 6-month timepoint as well as the calculate change images. 
Significant fixels colorized by Cohens D and thresholded to a p-value of p<0.05 are displayed on the whole brain tractogram and 
displayed in an axial lightbox view with 5mm seperation between slices. Yellow-red designates regions where the mean of the mTBI 
group is higher than controls and light to dark blue indicates regions where the mean for mTBi is lower than controls. Scaling for ease of 
visualization is from 0 to the maximum Cohen’s D with red and dark blue representing regions with maximum Cohen’s D. No Cohen’s 
D was greater than 0.2. For change images blue represents significant regions where mean change was lower in mTBI compared to 
controls and vice versa for red.  Images are presented in typical radiologic convention with the left side of the brain on the right.     

 85 

 
CHAPTER 4: A POPULATION-BASED SAMPLE OF MILD TRAUMATIC BRAIN INJURY 
REVEALS ABNORMAL WHITE MATTER STRUCTURE 6-MONTHS                              
POST-INJURY WITH FIXEL-BASED ANALYSIS: A TRACK-TBI STUDY 

After attempting the initial FBA in the sample in Chapter 3, the analysis was conducted with 

the whole sample to maximize power to detect differences between groups. All subjects included in 

the analysis in Chapter 3 were reanalyzed from their raw data with the rest of the sample that had 

multi-shell data using the adjusted analysis pipeline presented in this chapter.  The FBA extends the 

work that has been done to date with the TRACK-TBI diffusion data by applying a model that not 

only account for crossing fibers like High Angular Resolution Diffusion Imaging (HARDI), but 

additionally allows a comprehensive analysis of WM regions without the need for skeletonization 

with FSL’s TBSS. This should allow more precise localization of effects compared to the NODDI 

analysis conducted previously. Additionally, we sought to replicate the previous findings by the 

TRACK-TBI investigators in post hoc analyses with tensor models and NODDI to add important 

context to the FBA findings.  

INTRODUCTION 

Traumatic brain injuries categorized as mild account for the majority, up to 90%, of injuries 

worldwide, and are a major source of disability in a subset of patients with persistent 

symptoms.9,28,205,206 A bimodal distribution of prevalence exists, as well as trauma etiology, in which 

older individuals sustain TBI in falls, while younger individuals experience accidents, including 

motor vehicle accidents and sports-related mTBI. Of particular concern are the rising rates of death 

and disability from falls resulting in TBI in the aging US population, which is likely to increase based 

on current demographic and injury trends.30 Additionally, clinicians still rely on a thorough history 

and physical exam to categorize injuries, and recovery is defined based on the subjective resolution 

of symptoms and a return to “normal” functioning.1 There is a great need to develop neuroimaging 

 86 

 
 
 
biomarkers that can help clinicians prognosticate outcomes and allow the development of 

interventions that target the underlying pathophysiologic changes of mTBI as opposed to the 

symptoms. 

Conventional neuroimaging is not required to diagnose mTBI, and most often reveals no 

intracranial pathology.1,207,208 Despite that, persistent WM changes have been detected in the chronic 

stages of recovery following mTBI with DWI even beyond one-year post-injury, and changes in 

DTI metrics have been correlated to symptoms and functioning throughout the recovery period in 

the majority of studies.17,160,209,210 Still, however, several recent reviews have concluded that 

inconsistent patterns of change following mTBI are frequently observed with DTI, and weaknesses 

of DTI, the most widely used technique to date, have limited interpretation.3,17  

Contemporary DWI modelling techniques such as fixel-based analysis(FBA) have yielded 

interesting insights into pathologic conditions211,212 and WM development in children194,213, but only a 

few studies have applied it in mTBI.176,214,215 The largest study by Wallace and colleagues did not find 

any differences in the FBA metrics FD, FC, or FDC in comparisons of mTBI patients and controls; 

however, the wide range of injury to scan time may have contributed to this null finding (mean: 

194.4d± 38.3, range 98-338 days).215 A smaller study in amateur Australian football players that 

measured WM change in the acute (48hr) and subacute period (2 weeks) following injury found a 

greater number of symptoms and symptom severity in male athletes which is contrary to previous 

work. This occurred in the setting of significant sex differences in fixel-based metrics with males 

having increased FD in the cingulum and corticospinal tract compared to females.214 The most 

recent study in Australian football players that compared mTBI patients scanned in acute with those 

scanned in the subacute period found acute increases in FD, FC, and FDC.176 A review of these 

early studies shows that fixel-based metrics may be sensitive to a transient increase in axonal 

 87 

 
 
diameter due to post-injury cytotoxic edema in the acute period, that is followed by axonal 

degeneration in subacute and chronic periods reflected as decreased FD, FC and FDC.  

We demonstrate that contemporary methods can be applied to legacy datasets that 

reproduce original findings and extend our understanding of previously collected data. These data 

represent innumerable research hours and public research funds. The present study reinforces the 

importance of open data sharing in neuroimaging studies. To date TRACK-TBI studies have utilized 

the analysis technique TBSS. TBSS improves subject registration and avoids partial volume effects; 

however, by design, it limits the analysis to a subset of WM voxels  with relatively high FA 

representing only a fraction of all white matter regions.138,172,177 Additionally, recent calls that future 

studies should focus on multimodal methodology further limit analyses with TBSS, as the results in 

the skeletonized WM are not easily compatible with registration to other modalities.3 Additionally, in 

the present study, we conduct a comprehensive post hoc tract of interest analysis utilizing the highly 

reproducible deep-learning tracking algorithm TractSeg, which is trained on Human Connectome 

Project data and verified in several external samples.216  This approach is an improvement over 

traditional atlas-based approaches which involve an additional interpolation step.216,217 Within these 

tracts we present the relationship of tract averaged diffusion metrics estimated with DTI, DKI and 

NODDI between mTBI patients and controls for comparison with the findings of TRACK-TBI 

with these same techniques.120  

MATERIALS & METHODS 

Overview 

The TRACK-TBI Study is a multi-site clinical cohort study that involved 18 clinical sites and 

collected patient data across the spectrum of TBI, including mild, moderate, and severe.15,83,112 

Patients were enrolled on three paths: 1) evaluated and discharged from the emergency department 

(ED), 2) admitted to the hospital but not the ICU, or 3) patients admitted to the ICU. Additionally, 

 88 

 
300 extracranial trauma controls (orthopedic controls) were collected, 100 each from each 

enrollment path. The present study will focus on the CA+MRI sub-cohort, which includes 600 

participants.112 In addition, the CA+MRI group had anatomical MRI and DWI collected at 2-weeks 

and 6-months post-injury.113  

TRACK-TBI: Participants 

Participants had 1) a documented TBI based on the American College of Rehabilitation 

Medicine <24 hours before presenting to a site, 2) received a brain MRI/CT, 3) adequate visual 

acuity and hearing for testing, and 4) to be fluent in English or Spanish and provide consent. 

Excluded participants were those that met at least one of the following criteria: 1) had significant 

polytrauma present, 2) were prisoners, 3) were pregnant, 4) were on a psychiatric hold, 5) had a 

major baseline debilitating mental health disorder, 6) had a major debilitating neurologic disease, 7) 

had a significant history of preexisting health conditions, 8) had contraindications to MRI, 9) had a 

low-likelihood of follow up, 10) had a penetrating TBI, or 11) had spinal cord injury with ASIA 

score of C or worse. For the present study, we will focus on the mTBI patients with a presenting 

Glasgow coma scale (GCS) of 13-15 in the CA+MRI group (N= 554) who had multi-shell 

diffusion-weighted magnetic resonance imaging data collected (N= 86).  

TRACK-TBI: Neuroimaging  

Neuroimaging data were collected at 2-weeks and 6-months post-injury. Each participant 

underwent a scanning protocol that consisted of the following standard 3T MRI sequences: axial 

three-dimensional (3D) MPRAGE or IRFSPGR T1-weighted images (echo time [TE] = 1.5 ms; 

repetition time [TR]= 6.3 ms; inversion time = 400 ms; flip angle, 15 degrees) with 230 mm, field of 

view [FOV], 156 continuous slices (1.0 mm) with a 256 × 256 matrix size. Whole-brain DWI was 

performed ([TE] = 81 ms; [TR] = 9 s) using 64 diffusion-encoding directions, acquired at b = 1300 

s/mm2 and for the high definition fiber tracking (HDFT) group multi-shell data was collected (64 

 89 

directions at b = 1300 and 64 directions at 3000 s/mm2), eight acquisitions at b = 0 s/mm2 for each 

set of 64 directions, slices of 2.7-mm thickness each with no gap between slices, a 128 x128 matrix, 

and FOV of 350 x 350 mm.15,120 Ten participants did not pass quality check at several stages of the 

neuroimaging processing pipeline resulting in a final sample of (N=76).  

Image Preprocessing & Analysis  

All participant images were preprocessed using a combination of commands from FSL 

version 6.0 and MRtrix3 64-bit release version 3.0.3, and the analysis is schematically represented in 

Figure 4.1. Raw DWI images downloaded from FITBIR underwent denoising, Gibb’s ringing 

artifact removal, motion, and eddy current correction, and bias field correction. No susceptibility 

field distortion correction was conducted due to the lack of reverse-phase encoded b0 images for 

input to FSL’s topup tool. Quality assessment of all preprocessed images was conducted, and it was 

determined that data quality was reasonable without this correction. To facilitate the concatenation of 

the two separate acquisitions for use with MSMT- CSD, diffusion-weighted images were normalized 

by the mean of the b = 0 s/mm2 images to offset slight differences in signal intensity due to small 

differences in TE between the b=1300 and 3000 acquisitions. This was done in a way previously 

described by Chang & colleagues 2015.218 Participant masks for subsequent analysis steps were 

generated using the deep-learning optimized automated segmentation software FastSurfer and those 

generated with the dwi2mask were then coregistered to the mean b0 images of the DWI data.219 Masks 

with the best fit were selected, and all subject masks were visually inspected as an overlay on the 

preprocessed DWI data. When necessary, masks were eroded by one to two voxels to exclude any 

extra-axial voxels from subsequent analysis steps. A visual summary of the neuroimaging pipeline is 

presented in (Figure 4.1).  

 90 

 
 
 
 
Diffusion Tensor and Kurtosis Analysis 

Tensor models were fit with the functional magnetic imaging of the brain (FMRIB)’s diffusion 

toolbox, and the following tensor metrics were computed: FA, MD, AD, RD, KFA, MK, AK, and 

RK.220–222  

NODDI  

A multicompartment model was fit to the preprocessed DWI data with the NODDI toolbox 

(www.nitrc.org/projects/noddi_toolbox, Accessed Jun 2023) for MATLAB.134 The NODDI metrics 

computed were the ODI, NDI, and FWF.  

Fixel-Based Analysis    

Three tissue response functions were estimated from preprocessed DWI data using the 

Dhollander algorithm.199,200 A subset of control response functions was then averaged for the whole 

sample to create average tissue response functions for the WM, CSF, and GM. DWI images were 

then upsampled to an isotropic voxel size of 1.25 mm. Anatomical brain masks derived from 

Fastsurfer were registered to the DWI data and visually checked before the subsequent analysis 

steps. FODs functions were estimated using the single set of averaged response functions with the 

MSMT-CSD algorithm.193  Joint bias field correction and intensity normalization were then 

conducted on the FODs to approximate signal amplitude across subjects and correct for sample-

wide intensity differences not corrected in previous individual bias field correction preprocessing 

steps. 

 A whole sample unbiased template was generated  as described in Chapter 3. First, each 

individual subject's first and second timepoint FODs were brought into an intra-template space. All 

subject brain masks were then registered to this intra-subject template, and the intra-subject FODs 

and masks were used as input to the population template script a second time to generate a whole 

sample template.  

 91 

 
 
 
 
 
All subject FOD images from time points one and two were warped into the whole sample 

template space.202 All subject masks were then warped to the whole sample template using those 

same transformations and then averaged to create a whole sample mask. The whole sample mask in 

template space was visually inspected to ensure minimal extra-axial regions were included for 

subsequent analysis steps. A fixel mask was generated from the whole sample template FODs within 

the whole sample mask, defining the fixels that would be included in subsequent analysis steps. 

Subject fixels were then segmented from their warped FODs in whole sample template space. These 

subject fixels were then reoriented in whole sample template space and assigned to template fixels, 

specifying which fixels across subjects match fixels in the template space. FD, FC, and the product 

FDC were computed.203,204 The FC measure was log-transformed for statistical comparison. Change 

images calculated in a manner identical to that in Chapter 3 were computed. Briefly, longitudinal 

change images were calculated by subtracting the 2-week image from the 6-month image and 

dividing by time in years between scan dates as in Equation 3.1. 

Probabilistic tractography was conducted on the whole sample template FODs using 

MRtrix’s tkgen command.139 The resultant twenty million streamline tractogram was thinned to two 

million streamlines using SIFT filtering to account for false positive connections.223 This tractogram 

was then used to generate a fixel-fixel connectivity matrix to facilitate the connectivity-based fixel 

enhancement aspect of the statistical analysis.126 The fixel maps for FD, FC, and FDC were then 

smoothed using this matrix.     

Statistical Analyses 

Demographics 

Means and standard deviations for demographic variables were calculated. Normality was 

tested for all demographic variables. Due to the presence of outliers, non-normal distribution of the 

 92 

 
 
 
variable age, and the presence of low counts in some cells of contingency tables for categorical 

variables, non-parametric tests for independence were utilized.  

Whole Brain Fixel-Based Analysis  

 General linear models (GLM) comparing FD, logFC, and FDC across the groups were 

conducted cross-sectionally for the 2-week and 6-month timepoint images, and for the longitudinal 

change images. In addition to the family-wise error correction done implicitly a Bonferroni corrected 

alpha value of 0.0083 was used to designate significance. This alpha value reflects an additional 

correction for the multiple comparisons of the three metrics FD, logFC, and FDC, and the two 

directions tested in each comparison.224 The model was run with demeaned age, sex and handedness 

as potential confounding variables. Similarly, the log transformed intracranial volume was controlled 

for in the FC and FDC models.225 

Post-hoc Tract-wise Analysis 

Spherical harmonic peaks were extracted from the whole sample WM FOD template for 

input to TractSeg, an automatic WM segmentation algorithm that utilizes a convolutional neural 

network.216 The default 72 WM tracts created by TractSeg were calculated using the default 2000 

streamlines per tract. As these tracts were generated in template space, they were subsequently used 

to create binarized fixel and voxel masks for post-hoc tract-wise analysis with a method similar to as 

previously described at (https://github.com/smeisler/Meisler_Reading_FBA: Accessed Dec 

2023).213 Mean and standard deviation for FBA, NODDI, DKI, and DTI metrics were calculated 

within these masks for images from the 2-week and 6-month timepoints and compared between and 

within mTBI and controls groups with unpaired and paired two-sample t-tests respectively in 

MATLAB Version: 9.13.0.2126072 (R2022b) using the Statistical and Machine Learning Toolbox 

Version 12.4.226 Multiple comparison correction was conducted by applying false discovery rate 

 93 

 
 
(FDR) correction with the MATLAB Bioinformatics Toolbox Version 4.16.1 using the default 

Storey (2002) procedure, setting the threshold for significance to q < 0.05.227  

 94 

 
 
Figure 4.1- Schematic representation of standard preprocessing and analysis pipeline for fixel-based 
analysis adapted from Genc et al., 2018. This schematic includes the modified preprocessing in which 
we omitted correction for susceptibility field distortions and representative images generated 
throughout processing as examples. (Top Row Fifth Column)- average tissue response functions at 
each b-value for white matter (WM), grey matter (GM) and cerebrospinal fluid (CSF), (Second Row 
First Column)- green oval represents the left centrum semiovale a region known to contain crossing 
fibers, green box represents region of interest displayed to the left with WM FODs displayed, (Third 
Row Fourth column)- axial view of the whole sample WM FOD template generated from the linear 
and non-linear registration of all intra-timepoint subject templates, (fourth row second column)- whole 
brain tractogram with two-hundred thousand streamlines after processing with SIFT, (bottom row 
third column)- 72 tracts generated with TractSeg with isolated example tracts including the right and 
left corticospinal tracts and arcuate fasciculi and all streamlines traversing various the corpus callosum.  

 95 

 
 
 
 
RESULTS 

Sample Demographics 

No significant differences were observed between mTBI and controls for any demographic 

variables. Means and standard deviations for the whole sample, mTBI patient and control groups, 

and the associated p-values from the Wilcox rank sum and Fischer’s exact tests are reported in Table 

4.1. 

 96 

 
Table 4.1. Demographic Characteristics of Whole Sample, mTBI patients, and Controls 

Age (years) 
Gender 
Female 
Male 
Race 
Asian 
Black or African American 
Native Hawaiian or Other 
Pacific Islander 
Unknown 
White 
Ethnicity 
Hispanic or Latino 
Not Hispanic or Latino 
Right-Handed 
Both hands 
Left hand 
Right hand 
Years of Education 
10th Grade 
Associate’s degree 
Bachelor's degree 
Doctoral degree 
GED or equivalent 
High school graduate 
Master's degree 
Professional school degree 
Some college, no degree 
Unknown 
School Dropout 
Missing 
No 
Yes 
Expelled from School 
Missing 

Whole Sample 

N = 76 

37 ± 14 

23 (30%) 
53 (70%) 

13 (17%) 
5 (6.6%) 

1 (1.3%) 

2 (2.6%) 
55 (72%) 

15 (20%) 
61 (80%) 

3 (3.9%) 
5 (6.6%) 
68 (89%) 

1 (1.3%) 
4 (5.3%) 
29 (38%) 
3 (3.9%) 
1 (1.3%) 
4 (5.3%) 
13 (17%) 
2 (2.6%) 
17 (22%) 
2 (2.6%) 

2 (2.6%) 
71 (93%) 
3 (3.9%) 

mTBI 

N = 60 

36 ± 14 

16 (27%) 
44 (73%) 

9 (15%) 
2 (3.3%) 

1 (1.7%) 

2 (3.3%) 
46 (77%) 

12 (20%) 
48 (80%) 

1 (1.7%) 
5 (8.3%) 
54 (90%) 

1 (1.7%) 
3 (5.0%) 
23 (38%) 
2 (3.3%) 
1 (1.7%) 
4 (6.7%) 
9 (15%) 
2 (3.3%) 
13 (22%) 
2 (3.3%) 

2 (3.3%) 
56 (93%) 
2 (3.3%) 

2 (2.6%) 

2 (3.3%) 

 97 

Controls 

N = 16 

39 ± 14 

7 (44%) 
9 (56%) 

4 (25%) 
3 (19%) 

9 (56%) 

3 (19%) 
13 (81%) 

2 (13%) 

14 (88%) 

1 (6.3%) 
6 (38%) 
1 (6.3%) 

4 (25%) 

4 (25%) 

15 (93.7%) 
1 (6.3%) 

p-value 

0.418 
0.226 

0.145 

1 

0.122 

0.967 

0.704 

1 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 4.1. (cont’d)  

Whole Sample 

mTBI 

Controls 

p-value 

N = 76 
69 (91%) 
5 (6.6%) 

N = 60 
54 (90%) 
4 (6.7%) 

N = 16 
15 (93.7%) 
1 (6.3%) 

0.735 

1 (6.3%) 

1 (6.3%) 

13 (81%) 
1 (6.3%) 

2 (2.6%) 
1 (1.3%) 
3 (3.9%) 
3 (3.9%) 
50 (66%) 
2 (2.6%) 
3 (3.9%) 
1 (1.3%) 
2 (2.6%) 
2 (2.6%) 
1 (1.3%) 
6 (7.9%) 

2 (3.3%) 
1 (1.7%) 
2 (3.3%) 
3 (5.0%) 
37 (62%) 
1 (1.7%) 
3 (5.0%) 
1 (1.7%) 
2 (3.3%) 
1 (1.7%) 
1 (1.7%) 
6 (10%) 

No 
Yes 
Employed in Last 12 Months 
Missing 
1 Month 
10 Months 
11 Months 
12 Months 
3 Months 
4 Months 
5 Months 
6 Months 
8 Months 
9 Months 
N/A 
Insurance Status 
Missing 
Other 
Insurance purchased directly 
from an insurance company or 
on the health insurance 
exchange (this person or family 
member) 
Insurance through a current or 
former employer (of this person 
or another family member) 
Medicaid, Medical Assistance 
Medicare, 
Self-pay (Uninsured) 
TRICARE, VA or other 
military health care 
Table 4.1- Means with standard deviations and frequencies with percentages for demographic 
variables collected from mTBI participants and controls with multi-shell DWI data in the CA+MRI 
cohort of the TRACK-TBI study. P-values represent the Wilcox rank-sum test for independence 
and Fisher's exact test for independence, calculated using the ‘rstatix’ package in R Studio 
v2023.09.1+494.   

2 (3.3%) 
2 (3.3%) 
10 (17%) 

3 (3.9%) 
2 (2.6%) 
11 (14%) 

2 (3.3%) 
2 (3.3%) 

2 (2.6%) 
4 (5.3%) 

35 (58%) 

12 (75%) 

47 (62%) 

1 (1.7%) 

6 (7.9%) 

1 (1.3%) 

1 (6.3%) 

1 (6.3%) 

6 (10%) 

2 (13%) 

0.488 

 98 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Whole Brain FBA Results 

Compared to controls lower FDC in mTBI patients at the 2-week timepoint was primarily 

driven by differences in FD. Lower LogFC in the mTBI group in small regions of the posterior limb 

of the internal capsule at the 2-week timepoint remained at the 6-month timepoint and included 

frontal white matter regions in the later comparison Figures 4.6 and 4.7. Figure 4.2 shows fixels that 

had significantly decreased FDC in mTBI patients compared to controls at the 2-week timepoint; 

fixels are colored by Z-statistic and thresholded with family-wise error corrected p-values of p 

<0.0083. Significant differences were primarily localized to the anterior commissure, projection 

fibers including the left thalamo-prefrontal tract, internal capsule and genu of the corpus callosum. 

Figure 4.3 shows fixels that had significantly decreased FDC in mTBI patients compared to controls 

at 6-months post injury, fixels are colored by Z-statistic. Differences were observed in similar brain 

regions as the 2-week timepoint, with a greater number of significant fixels at 6-months. Primarily, 

differences were observed in the forceps minor fibers traversing the genu of the corpus callosum 

and projection fibers of the left thalamo-prefrontal tract. Smaller significant regions were also 

observed in the internal capsule, external capsule, and anterior commissure. No regions of increased 

FDC were observed at either timepoint, and no regions had significant differences in comparisons 

of change images. 

Significant differences in FD were observed throughout the WM especially in the deep WM 

at the 2-week timepoint as shown in Figure 4.4. These differences remained fairly stable over time 

with the majority of regions falling about 10-15% below the control mean at both timepoints. There 

were some regions of greater difference around 30%, including the anterior commissure and internal 

capsule, appearing at the 6-month timepoint (Figure 4.5). No regions with increased FD were 

observed, and no differences in FD change images survived family-wise error correction.  

 99 

 
 
Figure 4.2 – Axial, sagittal and coronal view of the white matter template with fixels from the fixel 
mask colorized by the GLM Z-statistic from comparisons of FDC at the 2-week timepoint post 
injury and thresholded with a family wise error corrected p-value of <0.0083.  

 100 

 
 
 
Figure 4.3 - Axial, sagittal and coronal view of the white matter template with fixels from the fixel 
mask colorized by the GLM Z-statistic from comparisons of FDC at the 6-month timepoint post 
injury and thresholded with a family wise error corrected p-value of <0.0083 

 101 

 
 
Figure 4.4 – Axial, sagittal and coronal view of the white matter template with fixels from the fixel 
mask colorized by the percent difference in mTBI compared to controls from comparisons of FD at 
the 2-week timepoint post injury and thresholded with a family wise error corrected p-value of 
<0.0083.  

 102 

 
 
Figure 4.5 – Axial, sagittal and coronal view of the white matter template with fixels from the fixel 
mask colorized by the percent difference in mTBI compared to controls from comparisons of FD at 
the 6-month timepoint post injury and thresholded with a family wise error corrected p-value of 
<0.0083 

 103 

 
 
 
Figure 4.6 – Axial, sagittal and coronal view of the white matter template with fixels from the fixel 
mask colorized by the GLM Z-statistic from comparisons of FC at the 2-week timepoint post injury 
and thresholded with a family wise error corrected p-value of <0.0083. 

 104 

 
 
 
Figure 4.7 – Axial, sagittal and coronal view of the white matter template with fixels from the fixel 
mask colorized by the GLM Z-statistic from comparisons of FDC at the 6-month timepoint post 
injury and thresholded with a family wise error corrected p-value of <0.0083.  

Post Hoc Tract-wise Results  

All post hoc comparison results from two-sample t-tests including the group mean, standard 

deviation at each timepoint, percent change, t-statistic and FDR corrected q-values are presented in 

Appendix Tables 1-14. Means, standard deviations and outliers are displayed as boxplots for tracts 

with significant comparisons at either the 2-week or 6-month timepoint are presented in Appendix 

Figures 1-94. Tract-wise results for significant metrics with direction of effect relative to controls for  

DTI, DKI, NODDI, and FBA are summarized for left and right projection and long association  

 105 

 
 
fibers, and commissural fibers in Figures 4.6-4.8. 

 Tensor Tract-wise Results 

 At the first timepoint mTBI patients had significantly higher tensor metrics AD, and AK 

throughout the brain in all projection fibers, long association fibers excluding the inferior 

longitudinal fasciculus, and commissural fibers. These differences remained at the second timepoint 

almost uniformly. Similarly, MD and MK were also elevated in widespread tracts throughout the 

brain in mTBI patients compared to controls at both timepoints. Far fewer tracts additionally 

exhibited increased RD and RK. No significant differences were observed in any tracts at either 

timepoint for the tensor metrics FA or KFA. No significant differences in within group 

comparisons between the first and second timepoint were observed for any tensor metrics Appendix 

Tables 15-22. 

NODDI and FBA Post Hoc Tract-wise Results 

A pattern of increased FWF, and decreased FDC, FD, ODI and NDI was observed in 

almost all WM tracts tested. Differences in FD, FDC, ODI and NDI were more commonly 

observed in left sided association fiber and projection fiber tracts Appendix Tables 1-14. The pattern 

of difference held uniformly throughout the brain at both timepoints. No tensor metrics compared 

within groups between the 2-week and 6-month timepoints were significant. The mTBI group had 

increased ODI and decreased FWF, NDI, FC and FDC at the second timepoint compared to the 

first Appendix Tables 23-28. The control group differed only in the FDC metric between the first 

and second timepoint with increased FDC at the second timepoint compared to the first.  

 106 

 
 
 
Figure 4.8-Summary of results from post hoc two-sample t-tests for DTI, DKI, NODDI, and FBA 
analyses for metrics estimated within projection fiber tracks with a significant q-value of q<0.05. 
The center of the image are the tract names with corresponding TractSeg generated tractograms. To 
the left and right are the significant metrics with the direction of the difference displayed with 
respect to controls. Metrics abbreviations: DTI: axial diffusivity (AD), mean diffusivity (MD), radial 
diffusivity (RD); DKI: axial kurtosis (AK), mean kurtosis (MK), radial kurtosis (RK); FBA: fiber 
density (FD), fiber density cross-section (FDC); NODDI: free water fraction (FWF), neurite density 
index (NDI), orientation dispersion index (ODI).  All metrics were significant at both timepoint 
unless otherwise indicated in brackets with the significant timepoint designated with P1 (2-week) or 
P2 (6-month). 

 107 

 
 
Figure 4.9-Summary of results from post hoc two-sample t-tests for DTI, DKI, NODDI and FBA 
analyses for metrics estimated within association fiber tracts with a significant q-value of q<0.05. 
The center of the image are the tract names with corresponding TractSeg generated tractograms to 
the left and right. Also on the left and right are the significant metrics with the direction of the 
difference displayed with respect to controls. Metrics abbreviations: DTI: axial diffusivity (AD), 
mean diffusivity (MD), radial diffusivity (RD); DKI: axial kurtosis (AK), mean kurtosis (MK), radial 
kurtosis (RK); FBA: fiber density (FD), fiber density cross-section (FDC); NODDI: free water 
fraction (FWF), neurite density index (NDI), orientation dispersion index (ODI). All metrics were 
significant at both timepoint unless otherwise indicated in brackets with the significant timepoint 
designated with P1 (2-week) or P2 (6-month). 

 108 

 
 
 
 
Figure 4.10-Summary of results from post hoc two-sample t-tests for 
DTI, DKI, NODDI and FBA analyses for metrics estimated within 
commissural fiber tracts with a significant q-value of q<0.05. The 
center of the image are the tract names with corresponding TractSeg 
generated tractograms to the left and right. Also, on the left and right 
are the significant metrics with the direction of the difference displayed 
with respect to controls. Metrics abbreviations: DTI: axial diffusivity 
(AD), mean diffusivity (MD), radial diffusivity (RD); DKI: axial 
kurtosis (AK), mean kurtosis (MK), radial kurtosis (RK); FBA: fiber 
density (FD), fiber density cross-section (FDC); NODDI: free water 
fraction (FWF), neurite density index (NDI), orientation dispersion 
index (ODI). 

 109 

 
 
 
 
 
Table 4.2 Summary of Findings in Chapter 4  
Main Finding 
1 

Compared to controls mTBI subjects had significantly lower fiber density, fiber 
cross-section and fiber density cross-section metrics from the fixel-based analysis 
at the two-week and six-month timepoints. Although difficult to know because no 
baseline or acute DWI images were collected my interpretation of the result is 
initially at the two-week timepoint lower FDC was driven by lower FD in the 
mTBI group compared to controls. In the context of the animal literature covered 
in Chapter 1 this finding reflects the axonal degeneration characteristic of the 
subacute period following injury. FC on the other hand is not significantly lower at 
this two-week subacute period likely due to the presence of microglia and 
astrocytes which play a significant role in the post-injury inflammatory response. 
At the six-month timepoint FC is likely now significantly lower in mTBI 
compared to controls due to the resolution of inflammation and consolidation of 
white matter tracts with fewer axons.  
 No comparisons of change images for the FBA were significant when comparing 
mTBI and control subjects. This demonstrates that there was not a significant 
difference in direction or rate of change in white matter characteristics between 
mTBI and controls. In the context of the animal and human literature covered in 
Chapter 1, this would seem to suggest that the mechanism of symptom resolution 
in mTBI patients during the subacute and chronic stages of recovery is unlikely to 
relate to a return of white matter microstructure towards control means which we 
consider normal or healthy in the present study.  
Post Hoc Comparisons revealed significant differences in tensor based metrics 
and metrics computed with higher order modeling techniques when comparing 
mTBI and controls. No differences for FA or KFA were observed for any white 
matter tracts, demonstrating the decreased sensitivity of these metrics in the 
presence of crossing fibers when averaged across an entire white matter tract 
mask. The observed pattern of differences reflects the same underlying 
pathophysiology described in Main Finding 1 above. mTBI patients had increased 
MD, MK, AD, AK, RD and RK in numerous white matter tracts reflecting greater 
isotropic diffusion throughout the brain due to the loss of tightly packed axons 
impeding water diffusion. This was reflected in higher order modelling metrics like 
increased FWF indicating greater isotropic diffusion, and decreased NDI and FD 
demonstrating decreased density of axons.  
Paired sample t-tests comparing group means within the mTBI group and control 
group between timepoints revealed no significant differences for any tensor 
metrics or fiber density. Increased ODI and decreased NDI and FWF were 
observed at the second timepoint compared to the first for the mTBI but not 
controls. Significantly increased FDC was observed for controls when comparing 
the first and second timepoint. In the absence of significant difference in FD or 
FC in controls, this is likely a false positive finding. This may be driven by outliers 
which could be further explored by either rerunning the comparisons with outliers 
removed or bootstrapping to estimate variance and assessing the stability of the t-
statistic.  

Main Finding 
2 

Main Finding 
3 

Main Finding 
4 

 110 

 
 
 
CHAPTER 5: FIXEL-BASED ANALYSIS IN A SAMPLE OF HIGH SCHOOL ATHLETES 
WITH SPORTS-RELATED MILD TRAUMATIC BRAIN INJURY 

INTRODUCTION 

The latest study to examine epidemiologic factors of traumatic brain injury in the United 

States was published in 2013. This study reported 640,000 emergency department visits for traumatic 

brain injury in children and adolescents under 14 years of age.228,229 In the context that the majority 

of traumatic brain injuries are categorized as mild, up to 90%, the report that 14% of children with 

mTBI will experience lifelong disability is alarming.229 A prominent etiology of traumatic brain injury 

among children and adolescents is sports-related concussion/mTBI. Similar to traumatic brain injury 

of other etiologies, diagnosis and return to play are based on resolution of symptoms and 

history/physical exam findings of phyicians.84 Accumulating evidence has demonstrated that the 

resolution of these physical signs and symptoms do not correspond to a return toward healthy 

control means of objective markers such as serum biomarkers and neuroimaging.24,125,170,230 

Additionally, there is evidence that even the stress and strain on the brain from a season of 

participation in contact sports can alter imaging markers including metrics of WM microstructure 

measured with DWI.156,231  

DTI has been the predominant tool utilized to study sport-related mTBI to date. However, a 

recent consensus has been reached that multimodal imaging studies are needed to fully understand 

the phenomena of resolving symptoms without resolution of observed changes in metrics like FA.3 

MSMT-CSD and FBA address the crossing fibers issue and account for partial volume effects at 

tissue interfaces, eliminating the need to skeletonize the diffusion metric images.193 Ultimately, the 

fiber-specific results are highly sensitive metrics that can be applied uniformly to the brain’s WM, 

which will facilitate the anatomical specificity needed to localize effects in the diffuse injuries 

characteristic of mTBI. 

 111 

 
 
 
 
During its initial development, the developers of MRtrix3 recommended a HARDI 

acquisition with high b-values of at least 2000 mm/s2 with at least 45 unique diffusion encoding 

directions. Later methodologic work in simulated and participant DWI data in childhood and 

adolescence showed increased sensitivity of higher b-value acquisitions when associating apparent 

fiber density and participant age.194  Still, other work has shown that even at low b-values (b=1000 

mm/s2) with fewer diffusion encoding directions (30), FD can be accurately quantified, albeit with 

an expected decrease in angular resolution.195 In this work, a comparison of CSD and DTI based 

deterministic and probabilistic tractography was compared in the corticospinal tract and arcuate 

fasciculus, along with comparisons of the precision of FD and FA estimation along the tract. It was 

shown that CSD improves on DTI-based tractography and provided more consistent estimation of 

FD along the tract. This work is important as it demonstrates that CSD is a clinically feasible 

technique that could be applied to estimate parameters with widely available acquisitions that require 

less data, a commonly cited strength of DTI. 

In the present study we sought to conduct an exploratory analysis of DWI data collected 

from adolescent athletes with mTBI. We used the contemporary analysis technique fixel-based 

analysis to do so. In this study, single shell data with a b-value of 1000 mm/s2 and 40 directions was 

collected without reverse phase-encoded images. We sought to explore the model fit on this data 

acquired in under 10 minutes with conventional preprocessing that would be applied in a typical 

DTI analysis. In the pursuit of study designs that are adequately powered factors such as sample size 

and scan time relate to cost per participant, and are important to consider to optimize use of public 

funds.232 In addition, when assessing the translational potential of a technique, the minimum 

necessary clinically feasible acquisition parameters that still allow adequate differentiation of lesions 

should be explored.233,234  

 112 

 
 
MATERIALS & METHODS  

Participants 

Participants were recruited from an affiliated concussion clinic at Michigan State University 

Health Care. Inclusion criteria for participants included: participation in high school athletics, age 

14-18 years, concussion in the last 8 days, and no metallic objects (such as dental braces) in the head 

and neck region. Ten participants completed all three timepoints, one participant completed only the 

first and second time points and one participant only completed the first timepoint.  This study was 

approved by the Michigan State University Institutional Review Board.  

Sport Concussion Assessment Tool 

Participant data was collected with a retrospective chart review in EPIC at Sparrow Health 

System. Responses from the administration of the Sport Concussion Assessment Tool 5th edition 

(SCAT-5) were compiled from their first visit to the concussion clinic and then again each 

subsequent visit they had until they were cleared to return to play.235 All available administrations of 

the SCAT-5 were plotted to examine the time from injury until the SCAT-5 symptom inventory was 

zero Figure 5.1. Creating these plots revealed three of eleven participants still had symptoms 30 days 

post-injury. 

Neuroimaging 

Twelve high school athletes completed MRI scans at an acute timepoint 5-8 days, subacute 

timepoint 1-month, and chronic timepoint 6-months post-injury on a GE 3T Signa® HDx MR 

scanner with an 8-channel head coil. T1-weighted images were collected for assessment of 

intracranial pathology with the following acquisition parameters: 184 1-mm sagittal slices, echo time 

[TE] = 3.8ms, repetition time [TR] of acquisition = 8.6ms, inversion time [TI] = 831ms, TR of 

inversion = 2332ms, flip angle = 8°, FOV = 250 mm ×250 mm, matrix size = 256×256, and 

 113 

 
 
 
 
receiver bandwidth = ± 20.8kHz. Whole-brain DWI was conducted employing a multislice single-

shot spin-echo EPI sequence. Key parameters included a [TE] set to minimum, repetition time [TR] 

= 12.8s, and 40 diffusion-encoding directions, acquired with a b-value of b= 1000 mm/s2, and four 

b= 0 s/mm2. Slices had a thickness of 2.4 mm with no inter-slice spacing and utilized a 128 × 128 

matrix with a field of view (FOV) of 220 mm. The patient was positioned supine with head-first 

entry with a total scan time of approximately 10 minutes. One incidental meningioma was 

discovered during the present study and the participant was ultimately included due to the extra-axial 

location of the pathology and a low likely hood of interference with WM.  

Image Preprocessing and Fixel-based Analysis 

Image preprocessing and analysis were identical to that described in Chapter 4, with a few 

notable exceptions. No susceptibility field distortion correction was conducted due to the lack of 

reverse-phase encoded b0 images. Similarly, due to this data being single-shell, the MSMT-CSD 

approach was applied in a similar fashion to that in Chapter 4; however, only the WM and CSF 

response functions were estimated, given that only one shell is available (b=1000). Finally, an 

additional template-building step was taken. First, WM FOD images from the 3-8 day and 1-month 

timepoint as well as the 1-month and 6-month were linearly and non-linearly coregistered. These 

resultant images were similarly linearly and non-linearly coregistered. These final intra-subject 

template images were coregistered for an unbiased whole sample population template.  

Statistical Analysis 

GLM were fit comparing FD, logFC, and FDC between the 3-8 day and 1-month 

timepoints, the 1-month and 6-month timepoints, and cross-sectionally comparing participants 

recovered within one month to those who took longer than one month. In addition to the family-

wise error correction done implicitly by the MRtrix3 fixelcfestats command, a Bonferroni corrected 

alpha value of 0.0083 was used to designate significance. This alpha value reflects correction for the 

 114 

multiple comparisons of the FD, logFC, and FDC, and the two directions tested in each 

comparison.  

RESULTS  

On the whole brain tractogram (Figure 5.1), significant differences at (p<0.01) before family-

wise error correction for multiple comparisons are displayed. The overlay is the whole brain 

tractogram colorized by effect size (Cohen’s D) from zero to the maximum effect size. The 

differences highlighted in yellow represent a positive effect in which the metric is higher in subjects 

with symptoms 30-days post-injury, and the differences highlighted in blue represent the opposite 

negative effect. Differences representing decreased FD, FC, and FDC are apparent peripherally in 

the cortical WM, and genu and body of the corpus callosum. Regions of increased FD, FC, and 

FDC are also present in the bilateral fornices and left inferior fronto-occipital fasciculus. The largest 

effects are in the deep WM, specifically in the anterior limb of the internal capsule, external capsule, 

and the forceps minor that persist to the  6-month timepoint when comparing participants 

symptoms after 30-days with the rest of the sample. No effects remained significant after family-

wise error correction. 

 115 

 
Figure 5.1- Results of the fixel-based analysis comparing a subsample of patients with a prolonged time-course for symptom resolution 
(red lines in top-left figure) compared to those with complete symptom resolution within one month post-injury.  Results are 
displayed in an axial lightbox view with 5mm spacing between slices, the whole sample population white matter FOD template as an 
underlay, and SIFT corrected whole brain tractogram with two hundred thousand streamlines colorized by Cohens D and 
thresholded with an image containing tract-wise uncorrected p-values of p<0.05. Panels are arranged in order of timepoint from the 
acute to chronic period (a-c).  

 116 

 
Figure 5.2- Results of the fixel-based analysis comparing group means between timepoints. Results 
are displayed in an axial lightbox view with 5mm spacing between slices, the whole sample 
population white matter FOD template as an underlay, and SIFT corrected whole brain tractogram 
with two hundred thousand streamlines colorized by Cohen’s D and thresholded with an image 
containing tract-wise uncorrected p-values of p<0.05. Panels are a comparison of the acute and 
subacute timepoints (left) and the subacute and chronic timepoint (right). 

Table 5.1 Summary of Findings in Chapter 5  

Main Finding 
1 

Main Finding 
2 

No significant differences between mTBI subjects when comparing within group between 
the first and second timepoints and second and third timepoints. Additionally, a cluster of 
subjects with symptoms one month post injury did not differ significantly in any fixel 
based metrics when comparing acute, subacute and chronic DWI scans.  
This data had significant susceptibility field distortions in the frontal and temporal regions 
in several subjects. This is evident in the frontal regions of the group average template. A 
reasonable next step would be application of Synb0 to each subject, segmentation of each 
individual subjects data from the acute, subacute and chronic timepoints with TractSeg 
and comparison of tract averaged metrics estimated in subject space with a less 
conservative correction for multiple comparisons like false discovery rate.  

 117 

 
 
 
 
 
 
 
CHAPTER 6: GENERAL DISCUSSION & FUTURE DIRECTIONS 

GENERAL DISCUSSION OF EXPERIMENTAL CHAPTERS 

Here, the findings from experimental Chapters 3-5 will be discussed in the context of the 

TRACK-TBI diffusion literature and the articles covered in Chapter 2. We also cover some of the 

pitfalls encountered in working with the open-access TRACK-TBI data from FITBIR. Finally, 

future directions for this work are outlined with short-term and long-term plans to accomplish them.   

The lack of significance in any FBA metrics in Chapter 3 is likely due to lack of power from 

the small sample size; the effect size of the diffuse injuries that occur in mTBI is small. The 

motivation for the small sample size in this initial analysis was to reserve the remaining participants 

with multi-shell data to facilitate replication of the analysis  in a second subgroup of the mTBI 

participants with multi-shell data similar to the analysis by Palacios and colleagues.120 Although we 

cannot know with complete confidence that it was a lack of power that led to our non-significant 

results in the pilot study, an interesting future direction would be to test with methods similar to that 

used in Genc et al., 2018 to see if the analysis was underpowered to detect differences with this 

sample size.197,236  Numerous small studies and large studies that have utilized DTI and 

contemporary modeling techniques have reported similarly negative results for some or all DTI 

metrics.17,122,161,174,191 However, Mito and colleagues noted significant increased in FDC and FD in 

their study of 29 Australian football players.176 Nonetheless, the results observed before family-wise 

error correction in the whole brain fixel-wise analysis follow a pattern that we would expect, with 

widespread diffuse difference, especially in the deep WM, like the internal and external capsule, and 

frontal WM regions, like the forceps minor fibers traversing the genu of the corpus callosum. It is 

well described that the effect size of mTBI injuries is quite small and requires adequate power to 

detect, especially for DTI metrics like FA.112 It remains to be seen if the added specificity from 

accounting for crossing fibers increases power in analyses with mTBI DWI data. 

 118 

 
 
 
Figure 6.1- Fractional anisotropy images from a single subject 
demonstrating the geometric distortions introduced with the 
use of Synb0-DISCO to generate undistorted b0 images for 
input to FSL’s topup. The top images are a sagittal view with 
the distorted image on the right, particularly evident in the 
brainstem region of this view. The bottom image is an axial 
view with emphasis on the genu of the corpus callosum where 
the geometric distortions are evident in this view. 

Figure 6.2- An MRI artifact, likely ghosting 
of the skull or less likely Gibbs ringing 
artifact, that was not removed in 
preprocessing. In subjects with this artifact, 
Synb0-DISCO attempted to interpolate data 
proximal to this arc and within the brain 
resulting in an anterior-posterior 
compression distortion of the data as in 
(Figure 6.1). 

 119 

 
 
 
 
 
 
 
We ultimately did not proceed with the initial pilot processing pipeline, as the use of Synb0-

Disco introduced severe geometric distortions in some patient data (Figure 6.1). Although CSD is an 

ill-posed problem and, therefore, highly susceptible to noise, after a quality check of the DWI data, it 

was deemed that the improvement in data quality would be only incremental for the subjects on 

which it was successfully implemented.22 After consulting with the originator of Synb0-Disco, Dr. 

Kurt Schilling, it was discovered that a scanner artifact present in some of the datasets was likely the 

culprit for the distortions introduced in the data (Figure 6.2).198 A forthcoming GitHub issue will be 

created to document this occurrence for future users of the TRACK-TBI dataset or those 

attempting to do susceptibility field distortion correction on open-access datasets in the future. 

Nonetheless, it is important to include this pilot pipeline, as theoretically, it should work on other 

data and would be a useful method to ensure optimal data quality in legacy datasets collected before 

the collection of reverse phase-encoded direction b0 images was standard. Finally, an important 

quality check in pipelines utilizing FBA and Synb0-DISCO is the visual inspection of images after 

distortion correction to detect potential artificial distortions.  

In Chapter 4 we demonstrate that FBA metrics of WM microstructure differ between mTBI 

and controls in the subacute and chronic periods following mTBI. The most prominent areas of 

difference in whole brain FBA were in the internal capsule, anterior regions of the corpus callosum, 

and projection fibers from the thalamus to anterior brain regions. Differences included more fixels 

and greater percent differences on the left side of the brain in both the whole brain FBA and post 

hoc analyses in this sample. Most interestingly, from the 2-week to 6-month timepoint there was not 

much resolution of injured tissue, which would have been indicated by a return of the mTBI mean 

towards the control mean and significant differences in the comparisons of longitudinal change 

images. This would seem to suggest that in this general population sample, the injured tissue 

characteristics did not change much in the subacute and chronic periods following injury.  

 120 

The TRACK-TBI investigators have previously applied the higher-order multicompartment 

modeling technique NODDI. In their analysis, they observed more widespread differences 

compared to the present analysis, however, a similar pattern of lower NDI was observed in both. 

NDI a measure of neurite density, which is similar to the fiber density metric of FBA. A possible 

explanation for this difference in localization could be their use of TBSS in the analysis which has 

been shown to generate artificially lower values of FA in prior studies.172 This highlights a major 

strength of the FBA method to comprehensively question all WM voxels while taking into account 

tract-specific changes with the connectivity-based fixel enhancement.19 Additionally, the NODDI 

TRACK-TBI study used false-discovery rate for correction of multiple comparisons which is a less 

conservative approach for correction compared to family-wise error correction. This is further 

supported by the widespread significant differences observed in the post hoc tract averaged 

comparisons of DTI, DKI and NODDI metrics in which we also utilized false discovery rate 

correction for multiple comparisons. Again, this highlights the challenge of multiple comparison 

correction in the field of neuroimaging. Given the current need to establish a consistent pattern of 

WM change as measured by DWI following injury, the field may have a higher tolerance for more 

false positives associated with false discovery rate. On the other hand, given the numerous factors 

that can result in DWI signal change at the microstructural level, such as extent of myelination, 

presence of other cell, and intra-axonal organelles to name a few, the more conservative family-wise 

error correction may in fact be preferred.105,237  

Prior findings with FBA in the literature have also been mixed largely due to variations in 

study design, which was part of the genesis of the TRACK-TBI study. For example, Wallace and 

colleagues 2020 were the first to conduct a FBA in a mTBI cohort, and found no differences in 

fiber-specific metrics.174 These findings are likely due to the high variability in scan time between 

subjects as discussed in Chapter 2, meaning WM at various stages of recovery were averaged, 

 121 

emphasizing the importance of consistent scan time in mTBI studies. In the TRACK-TBI NODDI 

study, for example, an injury to scan time standard deviation of 2-days and 8-days for the 2-week 

and 6-month timepoints respectively, was reported.112  

Only two other studies have utilized FBA in populations of sport-related mTBI patients. 

They both observed increased FDC primarily driven by increases in FD. These findings are in 

contrast to the analysis in this dissertation, which could be due to differences in WM reactivity to 

injury in athletes vs the general population. For example, it has been proposed that observations of 

WM change in contact sports athletes may reflect a synergistic effect of repetitive subconcussive 

blows accumulated during the season amplified in the setting of the trauma that resulted in the 

singular mTBI diagnosis.112,154,155,231 Similarly, in healthy athletic young populations the time course of 

pathophysiologic change as described in Chapter 1 may be different compared to the general 

population; this is an open question. It is likely, however, as the authors propose, this pattern reflects 

cytotoxic edema following injury, which may be in contrast to a predominant pattern of vasogenic 

edema and diffuse axonal injury followed by axonal degeneration in the general population.112,176,214  

It is unusual that the tensor metrics AD and AK were significantly different throughout the 

brain with no differences in FA or KFA. All of these metrics are related to the principal diffusion 

direction, which should be aligned along the principal fiber population in voxel. A possible 

explanation is that averaging across tracts with a large number of included crossing fiber regions may 

have artificially lowered FA and KFA values, however a comparable effect would occur in averaging 

AD and AK. Particularly in the context of concurrent observed decreases in NDI, FDC and FD 

which would suggest decreases in intraaxonal compartment signal and axonal loss, higher AD and 

AK in the mTBI group seems unlikely. Finally, increased FWF in numerous tracts at the six month 

timepoint is similarly difficult to explain as it is commonly observed to be increased in the acute 

phase of injury indicative of edema. This should be resolved in more chronic stages of recovery. It is 

 122 

more likely that this finding in conjunction with decreased FD, FDC and NDI reflect an increase in 

extracellular fluid space due to chronic axonal degeneration. This further demonstrates how 

multicompartment models offer improved interpretability of DWI results compared to traditional 

tensor metrics.  

There are several important limitations to the present study that should be considered in the 

interpretation of the results. First, the cohort was smaller than the entire TRACK-TBI CA+MRI 

cohort, owing to the availability of multi-shell data in only a small proportion of the TRACK-TBI 

sample. Multiband MRI technology, which accelerates the acquisition of data, was used widely in 

research in the early 2010s but was not widely available on clinical scanners until later. This may 

have motivated the primarily single shell acquisitions used for the TRACK-TBI study.238 

Furthermore, a goal of the TRACK-TBI study was to produce standardized and widely reproducible 

outcome measures for the study of TBI, and the novelty of multi-shell data and advanced modeling 

techniques may not have aligned with that goal. This is also a relative strength of this study, as all 

participants were collected at the same study site on the same scanner. There was also a difference 

between the number of participants in the control and mTBI groups. Therefore, the group means 

for WM from mTBI patients may have been estimated more precisely in this sample compared to 

the control group. In light of this small control group sample size, the t-test results in particular 

should be interpreted with caution. 

We observed no differences in FC in this study at both timepoints. This could be due in part 

to the lack of susceptibility field distortion correction. The construction of the population template 

and individual subject warps used to calculate FC are sensitive to distortions. Therefore, the lack of 

susceptibility field distortions was likely to have affected the quality of the metric estimation Chapter 

4 & 5. The work around is correction with Synb0-Disco, which was unsuccessful for the TRACK-

TBI data and visual inspection showed susceptibility field distortions were minor in subjects 

 123 

included in this analysis. However, this may be an important future direction for the data presented 

in Chapter 5.  

Although the two samples were well matched on demographic variables, WM is known to 

change with age, as well as other factors which were not controlled for in the present analysis.178 An 

important future direction will be fitting additional models that include demeaned age, sex, and 

handedness as covariates to confirm the present findings. Even then, there could be additional 

factors driving the differences between groups, including components of resilience to WM injury 

that have yet to be fully described. For example, in the TRACK-TBI NODDI analysis two clusters 

of mTBI patients differing on performance on neuropsychological test of processing speed and 

verbal memory were identified.112 These lower performing patients had, on average, lower years of 

education and had lower ODI in the central WM. These patients reported fewer symptoms than the 

higher performing group, suggesting components of resilience to injury may be contributing to 

differences in recovery.112  

As an emerging technique, FBA is continuing to evolve. Recently, the developers released 

guidelines on the correction of metrics for intracranial volume, and an additional next step will be 

adjusting for this with the results of a Freesurfer segmentation per the developer's 

recommendations.225 Finally, multiple comparisons are inherent to the high dimensional data in the 

field of neuroimaging and the biological significance of findings in this context can be challenging.239 

In this analysis, we perform conservative “strong” family-wise error correction by additionally 

correcting the alpha value for multiple comparisons to assign significance. However, this is not the 

method recommended by the MRtrix3 developers and may have reduced our sensitivity in this 

analysis.224    

Finally, with relation to CSD a potential limitation is the assumption of a single tissue 

response function for all WM. This is a core assumption of CSD and may simply be an inaccurate 

 124 

way to model WM. However, this issue is inherent to any modeling approach. In fact, in mild 

traumatic brain injury research, it could potentially be seen as a strength, as any fixel-wise deviations 

in WM could be attributed to partial volume effects as a result of the injury or the presence of 

multiple fiber orientations. Computing the average tissue response functions from the control group 

facilitates this, however, it is not currently a specific recommendation of the developers. A potential 

future analysis with an average response function from the whole sample or the mTBI patients alone 

could help add context to these findings. Along these lines, the test-retest reliability of MSMT-CSD 

has only been tested in data up to 3-months past a baseline scan.196 In the 3-month cohort of this 

study, the interclass correlation coefficient was good (ICC>0.8). However, the sample size was 

small, and more work is needed to understand the agreement of this measure over time, especially at 

longitudinal time points like the 6-month collection in TRACK-TBI.  

The animal literature covered in Chapter 1 lends important context to the findings in 

Chapter 5. We know based on the work in animals that the energy crisis as a result of the injury lasts 

well into the subacute period following injury, and contributes to cell death due to calcium 

toxicity.33,35,36 This would seem to support the observation that higher order modeling techniques 

that infer density of axons in a voxel were lower in mTBI patients compared to controls at the two-

week timepoint due to continued cell death and degeneration of axons during the subacute period. 

Similarly, fiber cross-section was not lower at this timepoint as the damaged tissue is likely 

undergoing astrogliosis and has an increased presence of microglia and non-resident inflammatory 

cells extravasated from the damaged BBB maintaining the cross sectional area of the tracts.48, 49, 50, 52, 

53 Along the same lines this increased cellularity would impede diffusion and is reflected by reduced 

MD, MK, AD, AK, RD, and RK in numerous tracts. These processes have likely resolved by the 6-

month timepoint as reflected by the increased regions of lower ODI and FC at this chronic stage of 

recovery. 

 125 

In Chapter 5 we take an exploratory look at the Michigan State University (MSU) Health 

Care concussion clinic participants with data collected before the COVID-19 pandemic. In this 

experiment, the general gestalt of Figure 5.1 is an expected pattern of decreasing differences in 

group means between participants with prolonged symptom recovery and those with a more typical 

course of symptom recovery. Notably, in the group mean comparisons between the acute and 

subacute periods post-injury shown in Figure 5.2, there is only a single right-sided frontal WM 

region where FDC was observed to be lower in the acute period following injury. This could 

indicate that a similar pattern of only modest changes occurs in WM during the acute to subacute 

recovery period. This is an imperfect interpretation however, and the true comparison will come 

with calculation and comparison of change images using the same approach as described in Chapters 

3 & 4. Similarly, the small sample size in comparisons and uncorrected results are likely heavily 

influenced by type-I errors. Further, these images were not corrected for susceptibility distortions, 

which may have affected the quality of the population template and, therefore, the subjects’ warps 

and fiber-cross section measure. An important next step will be attempting to apply the Synb0-

DISCO distortion correction to these data as in Chapter 3 and repeating the analysis with the 

recommended additional distortion corrections.18 This would allow application of TractSeg to each 

individual subject for direct estimation of metrics in subject native space. 

In the MSU Health Care study, more information about the remainder of the athletic 

seasons of participants were not collected. In an ideal study design, given the literature reviewed in 

Chapter 2, it is necessary to understand the impact exposure of high school athletes, and the length 

of time they have been exposed to contact practice pre-injury and after return to play that coincides 

with data collection periods.176,214 Additionally, this sample was entirely male, and highlights the 

underrepresentation of female mTBI participants in studies of sport-related concussion as identified 

in Chapter 2. We almost recruited a single female participant soccer player; however, she was 

 126 

scheduled to have dental braces installed during the course of the study. In future work, increased 

effort should be placed to recruit female participants, which should include targeted community 

engagement in the sports in which females experience increased head impact exposure such as 

soccer and basketball.25 This underscores the importance of future work to take into consideration 

the early descriptions in the literature of disparities in outcome, referral for clinical care, and follow-

up care adherence.240–242 

With respect to the single shell low b-value acquisition, Genc and colleagues 2020 

demonstrated that high b-value single-shell acquisitions applied with the single-shell multi-tissue 

CSD approach were more sensitive than multi-shell acquisitions.194  The interpretation being that 

isolating signal from the intracellular compartment with high-b value acquisitions in studies that are 

solely interested in observing differences in FD could be a viable strategy in mTBI. This underscores 

the importance of close collaboration with an MR Physicist who can guide researchers when 

selecting acquisitions that are specifically tuned to their research question.  

FUTURE DIRECTIONS 

In the future, it will be important to assess if metrics derived from DWI can be used to train 

predictive models of patient outcome following mTBI. For example, a support vector machine 

algorithm to classify DWI data in mTBI patients who did not recover by 6-months would be an 

interesting next step. Support vector machines (SVM) are supervised machine learning algorithms that 

learn a linear discriminant function.243 The goal of applying a SVM classification is to find a hyperplane 

in high-dimensional space between groups of neuroimaging data that predict a discrete class, in this 

case, recovery status at 6-months as indicated by GOSE total score. A similar approach, logistic 

regression, was successfully applied in the TRACK-TBI NODDI analysis.112 These methods are an 

improvement over the univariate methods in the present dissertation, as it allows the identification of 

regional patterns of pathology in disease, which lends itself to the current problem of poor localization 

 127 

 
in mTBI. Injured WM following an mTBI is not necessarily constrained to a specific ROI or single 

pattern of change; therefore, more data-driven modeling approaches may have greater success in 

predicting which patients will not recover. Compared to unsupervised approaches which may produce 

superior classifiers, SVM would allow examination of the factors that uphold the support vectors for 

increased understanding of what features the model is using for classification. 

We  would  expect  SVM  to  identify  patterns  of  regional  change  in  DWI  metrics  that  predict 

recovery with high sensitivity and specificity. A predictive DWI model would likely include WM features 

of decreased FA and increased MD in the corpus callosum, longitudinal fasciculi, internal capsule, and 

corona radiata as previously described in the literature.5,17 It would be interesting to compare if AD alone 

is more predictive than FA and MD, as this metric was more sensitive to differences in the post hoc 

tract-wise  analysis  conducted  in  this  dissertation.  We  would  expect  that  in  comparisons  of  models, 

including, tensor models and higher order modelling techniques such as FBA and NODDI, higher order 

modelling techniques would outperform tensor models. Additionally, it would be interesting to compare 

the performance of a model containing only clinical variables and FD, which seems to be the most 

sensitive metrics requiring the least data for estimation compared to all other metrics.194,195 We could use 

the tract-wise averages for the seventy-two tracts from TractSeg to train a SVM. Nevertheless, a superior 

approach  would  be  to  apply  TractSeg  to  each  individual  participant  and  estimate  the  diffusion 

parameters in native subject space. This would entail re-running TractSeg on each subject’s three primary 

spherical harmonic peaks extracted from their WM FODs. This would facilitate calculation of tract-wise 

metrics in native subject space for training the model as opposed to the current data analysis scheme 

which  includes  an  additional  interpolation  step  to  estimate  metrics  in  the  study  specific  WM  FOD 

template space. 

To add clinical context to the neuroimaging findings presented in this dissertation, we plan to 

explore subgroup differences in mTBI participants on the GOSE. In the absence of gold standard 

 128 

objective biological markers of outcome, clinical assessment tools such as the GOSE, are essential for 

understanding  how  observed  group  differences  in  WM  relate  to  patient  functioning  and  clinical 

outcomes. As a part of our collaborative agreement with TRACK-TBI, the present results from the 

FBA and post hoc tract-wise analysis will be shared with the TRACK-TBI biostatistics core to examine 

how metrics derived with this technique relate to outcome on the GOSE, symptoms on the RPQ, and 

verbal memory and processing speed on the Rey-Auditory verbal learning test.244 We will also request 

more detailed demographic information be computed on mechanism of injury. This important future 

direction will add context to the reductions in FD is related to differences in functioning in this sample. 

This is particularly important as many of the lasting effects of mTBI are related to post-injury deficits 

in function at work, in school, and in patients personal relationships.7,8,112    

An emerging objective marker of mTBI pathophysiology in human samples is alterations in 

cerebral blood flow and functional connectivity. Early work by Zhu and colleagues demonstrated a 

transient decrease in default mode network connectivity at 1-month post-injury. The authors 

proposed this distributed network is likely to be affected in a large proportion of diffuse mTBI 

injuries.125 These findings must be re-examined in the context of other modalities that quantify 

perfusion to confirm if the effect is related to disconnection or dysfunction on the neuronal side or 

changes in blood flow on the vascular side. ASL is an MRI technique that magnetically “marks” 

protons in arterial blood.245 When these protons reach a new vascular bed, they can be measured 

because of their unique magnetic signal. Early MRI studies in rats with induced TBI demonstrated 

that cerebral blood flow as measured by ASL was reduced acutely seven days after injury and even 

one year after injury compared to sham group.246,247 It is known that a biphasic change in nitric 

oxide, a potent vasodilator, and nitric oxide synthase, an enzyme that catalyzes the synthesis of nitric 

oxide, are in part responsible for the observed changes in cerebral blood flow.248 However, animal 

 129 

studies have demonstrated that in most cases, nitric oxide production and nitric oxide synthase 

expression return to pre-injury levels after approximately 7-10 days. 

Following mTBI, ASL studies in humans have revealed that there is a similar pattern of 

changes in cerebral blood flow as observed in animal studies. A recent study demonstrated a 

significant decrease in cerebral blood flow from 24-hours to 8-days post-injury in a human sample.47 

Importantly, scores on neuropsychological tests had returned to baseline levels at 8 days. The 

authors proposed this transient alteration in perfusion represents a window of cerebral vulnerability 

that current clinical management cannot assess. This is of particular importance when considering 

the diagnosis and return to play considerations around sports-related concussion. Work by Meier 

and colleagues, 2015 demonstrated similarly depressed cerebral blood flow at 1-day and 1-week 

post-mTBI when compared to age and fitness matched controls.249 This study also demonstrated a 

region-specific decrease in cerebral blood flow between participants who returned to play before 2-

weeks vs. after 2-weeks. As is common in early studies in the mTBI literature, a major limitation of 

sample size and variations in measurement timing temper the strength of this evidence. However, 

these early works demonstrate the potential detrimental effects of TBI classification that is not 

biologically performed with objective markers of neurophysiology. These early studies are 

compelling and demonstrate the importance of considering the time course of disturbed perfusion 

when utilizing imaging modalities dependent on the blood-oxygen-level-dependent signal in mTBI. 

A very unique study recently demonstrated that a wearable brain vibration device that may mimic 

forces experienced in sport or operation of machinery caused localized decreases in cerebral blood 

flow in the region of the posterior cingulate cortex and precuneus, both of which are hubs of the 

default mode network.250  

Given this early evidence, there is a great need to better understand the natural history of 

cerebral blood flow changes following mTBI. The TRACK-TBI study collected ASL data in a subset 

 130 

of patients, and it would be an interesting future direction to examine the localization of decreased 

perfusion with changes in resting-state fMRI functional connectivity at the subacute 2-week 

timepoint. Additionally, the high quality tractography results demonstrated in this dissertation would 

facilitate comparisons of structural and functional connectivity in participants with multi-shell data 

and resting-state fMRI data that could be colocalized with alterations in perfusion found with ASL. 

ASL findings from the department of defense concussion research and education (CARE) study 

demonstrated alterations in cerebral blood flow acutely following injury in sports-related mTBI.251 

Most recently, studies of cerebral blood flow alterations in non-injured contact sport athletes 

demonstrated alterations in pre and post-season ASL scans lending increased strong evidence to this 

line of investigation.252 Finally a recent review of ASL findings in 23 mTBI studies concluded there is 

strong evidence for a relationship between alterations in ASL measured cerebral blood flow and 

clinical recovery.253 

In the pursuit of discriminative technologies that will not overly burden the already high 

healthcare costs in the US, it is important to consider cost-effectiveness in clinical research.254 

Magnetic resonance imaging is an expensive technology, and requires in person appointments for 

the patients, personnel to run the scanner, and physicians to interpret the images. Functional near 

infrared spectroscopy (fNIRs), measures alterations in cerebral blood flow in peripheral brain 

regions using light which can penetrate the skull and detect alterations in oxygenated and 

deoxygenated hemoglobin based on light absorption.255 Although less sensitive than MRI, fNIRs is a 

simple wearable technology. A recent study demonstrated a functioning device can be constructed 

for as low as $215.256 An intriguing early study in mTBI, demonstrated decreased brain oxygenation, 

as measured by fNIRs, in patients suffering from persistent post-concussive symptoms.257 More 

recently, acute alterations in cerebral blood flow and oxygenation have been observed with fNIRs in 

mixed martial arts athletes between pre to post contact training session scans but not non-contact 

 131 

training sessions.258 These findings were significantly associated with cumulative angular 

accelerations as measured by kinematic mouth pieces. Future studies could benefit from using this 

technology to study the natural history of cerebral blood flow in an emergent setting and 

longitudinally with serial measurement at regular intervals during the recovery period. Presumably, 

patients could even be trained to take a device home for real time monitoring of alterations by a 

physician, which further necessitate greater investigation with this device. 

Conclusion  

Taken together, DWI is sensitive to even the diffuse minor WM changes in patients with 

mTBI of varying location and mechanism in the general population. Contemporary modelling 

techniques facilitate better interpretability of results by assigning effects to specific fiber populations 

within a voxel in the case of FBA, or to specific pathophysiologic mechanisms such as edema as in 

the NODDI metric FWF. Additionally, the use of reproducible tracking algorithms such as TractSeg 

are an improvement over atlas-based methods, allowing comprehensive localization of effects as 

measured by DWI in the whole brain. Albeit not ideal, tract averaging of metrics is more reliable 

when used in conjunction with methods that account for crossing fibers. Ultimately, such methods 

will allow the colocalization of effects observed with other modalities to fully describe the natural 

history of brain change following mTBI.

 132 

REFERENCES 

1.  Silverberg ND, Iverson GL, Cogan A, et al. The American Congress of Rehabilitation Medicine 
Diagnostic Criteria for Mild Traumatic Brain Injury. Arch Phys Med Rehabil. 2023;104(8):1343-
1355. doi:10.1016/j.apmr.2023.03.036 

2.  Georges A, M Das J. Traumatic Brain Injury. In: StatPearls. StatPearls Publishing; 2024. 
Accessed February 29, 2024. http://www.ncbi.nlm.nih.gov/books/NBK459300/ 

3.  Puig J, Ellis MJ, Kornelsen J, et al. Magnetic Resonance Imaging Biomarkers of Brain 

Connectivity in Predicting Outcome after Mild Traumatic Brain Injury: A Systematic Review. J 
Neurotrauma. 2020;37(16):1761-1776. doi:10.1089/neu.2019.6623 

4.  Kim HJ, Tsao JW, Stanfill AG. The current state of biomarkers of mild traumatic brain injury. 

JCI Insight. 3(1):e97105. doi:10.1172/jci.insight.97105 

5.  Asken BM, DeKosky ST, Clugston JR, Jaffee MS, Bauer RM. Diffusion tensor imaging (DTI) 
findings in adult civilian, military, and sport-related mild traumatic brain injury (mTBI): a 
systematic critical review. Brain Imaging Behav. 2018;12(2):585-612. doi:10.1007/s11682-017-
9708-9 

6.  McMahon PJ, Hricik A, Yue JK, et al. Symptomatology and Functional Outcome in Mild 
Traumatic Brain Injury: Results from the Prospective TRACK-TBI Study. J Neurotrauma. 
2014;31(1):26-33. doi:10.1089/neu.2013.2984 

7.  Nelson LD, Temkin NR, Dikmen S, et al. Recovery After Mild Traumatic Brain Injury in 
Patients Presenting to US Level I Trauma Centers: A Transforming Research and Clinical 
Knowledge in Traumatic Brain Injury (TRACK-TBI) Study. JAMA Neurol. 2019;76(9):1049. 
doi:10.1001/jamaneurol.2019.1313 

8.  Machamer J, Temkin N, Dikmen S, et al. Symptom Frequency and Persistence in the First Year 
after Traumatic Brain Injury: A TRACK-TBI Study. J Neurotrauma. 2022;39(5-6):358-370. 
doi:10.1089/neu.2021.0348 

9.  Cassidy JD, Carroll L, Peloso P, et al. Incidence, risk factors and prevention of mild traumatic 
brain injury: results of the who collaborating centre task force on mild traumatic brain injury. J 
Rehabil Med. 2004;36(0):28-60. doi:10.1080/16501960410023732 

10.  Wilde EA, Wanner IB, Kenney K, et al. A Framework to Advance Biomarker Development in 
the Diagnosis, Outcome Prediction, and Treatment of Traumatic Brain Injury. J Neurotrauma. 
2022;39(7-8):436-457. doi:10.1089/neu.2021.0099 

11.  Bigler ED. Neuroimaging Biomarkers in Mild Traumatic Brain Injury (mTBI). Neuropsychol Rev. 

2013;23(3):169-209. doi:10.1007/s11065-013-9237-2 

12.  Gan ZS, Stein SC, Swanson R, et al. Blood Biomarkers for Traumatic Brain Injury: A 

Quantitative Assessment of Diagnostic and Prognostic Accuracy. Front Neurol. 2019;10:446. 
doi:10.3389/fneur.2019.00446 

 133 

 
13.  Visser K, Koggel M, Blaauw J, Van Der Horn HJ, Jacobs B, Van Der Naalt J. Blood-based 

biomarkers of inflammation in mild traumatic brain injury: A systematic review. Neurosci Biobehav 
Rev. 2022;132:154-168. doi:10.1016/j.neubiorev.2021.11.036 

14.  Silverberg ND, Gardner AJ, Brubacher JR, Panenka WJ, Li JJ, Iverson GL. Systematic Review 

of Multivariable Prognostic Models for Mild Traumatic Brain Injury. J Neurotrauma. 
2015;32(8):517-526. doi:10.1089/neu.2014.3600 

15.  Yuh EL, Cooper SR, Mukherjee P, et al. Diffusion Tensor Imaging for Outcome Prediction in 
Mild Traumatic Brain Injury: A TRACK-TBI Study. J Neurotrauma. 2014;31(17):1457-1477. 
doi:10.1089/neu.2013.3171 

16.  O’Donnell LJ, Westin CF. An Introduction to Diffusion Tensor Image Analysis. Neurosurg Clin 

N Am. 2011;22(2):185-196. doi:10.1016/j.nec.2010.12.004 

17.  Lindsey HM, Hodges CB, Greer KM, Wilde EA, Merkley TL. Diffusion-Weighted Imaging in 

Mild Traumatic Brain Injury: A Systematic Review of the Literature. Neuropsychol Rev. 
2023;33(1):42-121. doi:10.1007/s11065-021-09485-5 

18.  Tournier JD, Smith R, Raffelt D, et al. MRtrix3: A fast, flexible and open software framework 

for medical image processing and visualisation. NeuroImage. 2019;202:116137. 
doi:10.1016/j.neuroimage.2019.116137 

19.  Raffelt DA, Smith RE, Ridgway GR, et al. Connectivity-based fixel enhancement: Whole-brain 
statistical analysis of diffusion MRI measures in the presence of crossing fibres. NeuroImage. 
2015;117:40-55. doi:10.1016/j.neuroimage.2015.05.039 

20.  Tournier J, Mori S, Leemans A. Diffusion tensor imaging and beyond. Magn Reson Med. 

2011;65(6):1532-1556. doi:10.1002/mrm.22924 

21.  Farquharson S, Tournier JD, Calamante F, et al. White matter fiber tractography: why we need 

to move beyond DTI: Clinical article. J Neurosurg. 2013;118(6):1367-1377. 
doi:10.3171/2013.2.JNS121294 

22.  Tournier JD, Calamante F, Gadian DG, Connelly A. Direct estimation of the fiber orientation 
density function from diffusion-weighted MRI data using spherical deconvolution. NeuroImage. 
2004;23(3):1176-1185. doi:10.1016/j.neuroimage.2004.07.037 

23.  Mccrory P, Meeuwisse W, Dvorak J, et al. Consensus statement on concussion in sport-the 5 th 

international conference on concussion in sport held in Berlin, October 2016 Consensus 
statement. Br J Sports Med. 2018;51:838-847. doi:10.1136/bjsports-2017-097699 

24.  Giza C, Greco T, Prins ML. Concussion: pathophysiology and clinical translation. Handb Clin 

Neurol. 2018;158:51-61. doi:10.1016/B978-0-444-63954-7.00006-9 

25.  Harmon KG, Drezner JA, Gammons M, et al. American Medical Society for Sports Medicine 

position statement: concussion in sport. Br J Sports Med. 2013;47(1):15-26. doi:10.1136/bjsports-
2012-091941 

 134 

26.  Sussman ES, Pendharkar AV, Ho AL, Ghajar J. Mild traumatic brain injury and concussion: 

terminology and classification. In: Handbook of Clinical Neurology. Vol 158. Elsevier; 2018:21-24. 
doi:10.1016/B978-0-444-63954-7.00003-3 

27.  Maas AIR, Menon DK, Steyerberg EW, et al. Collaborative European NeuroTrauma 

Effectiveness Research in Traumatic Brain Injury (CENTER-TBI): A Prospective Longitudinal 
Observational Study. Neurosurgery. 2015;76(1):67-79. doi:10.3174/ajnr.A1051 

28.  Frieden TR, Houry D, Baldwin G. The Report to Congress on Traumatic Brain Injury in the 

United States: Epidemiology and Rehabilitation. 
https://www.cdc.gov/traumaticbraininjury/pdf/tbi_report_to_congress_epi_and_rehab-a.pdf 

29.  Faul M, Coronado V. Epidemiology of traumatic brain injury. In: Handbook of Clinical Neurology. 

Vol 127. Elsevier; 2015:3-13. doi:10.1016/B978-0-444-52892-6.00001-5 

30.  Santos-Lozada AR. Trends in Deaths From Falls Among Adults Aged 65 Years or Older in the 

US, 1999-2020. JAMA. 2023;329(18):1605. doi:10.1001/jama.2023.3054 

31.  Katayama Y, Becker DP, Tamura T, Hovda DA. Massive increases in extracellular potassium 
and the indiscriminate release of glutamate following concussive brain injury. J Neurosurg. 
2009;73(6):889-900. doi:10.3171/jns.1990.73.6.0889 

32.  Takahashi H, Manaka S, Sano K. Changes in extracellular potassium concentration in cortex and 

brain stem during the acute phase of experimental closed head injury. J Neurosurg. 
2009;55(5):708-717. doi:10.3171/jns.1981.55.5.0708 

33.  Giza CC, Hovda DA. The new neurometabolic cascade of concussion. Neurosurgery. Published 

online 2014. doi:10.1227/NEU.0000000000000505 

34.  Hovda DA, Yoshino A, Kawamata T, Katayama Y, Fineman I, Becker DP. The increase in local 
cerebral glucose utilization following fluid percussion brain injury is prevented with kynurenic 
acid and is associated with an increase in calcium. Acta Neurochir Suppl (Wien). 1990;51:331-333. 

35.  Thomas S, Prins ML, Samii M, Hovda DA. Cerebral Metabolic Response to Traumatic Brain 
Injury Sustained Early in Development: A 2-Deoxy-D-Glucose Autoradiographic Study. J 
Neurotrauma. 2002;17(8):649-665. doi:10.1089/089771500415409 

36.  Yoshino A, Hovda DA, Kawamata T, Katayama Y, Becker DP. Dynamic changes in local 
cerebral glucose utilization following cerebral concussion in rats: evidence of a hyper- and 
subsequent hypometabolic state. Brain Res. 1991;561(1):106-119. doi:10.1016/0006-
8993(91)90755-K 

37.  Giza CC, Hovda DA. The New Neurometabolic Cascade of Concussion. Neurosurgery. 

2014;75(Supplement 4):S24-S33. doi:10.1227/NEU.0000000000000505 

38.  Giza CC, Maria NSS, Hovda DA. N -Methyl-D-Aspartate Receptor Subunit Changes after 

Traumatic Injury to the Developing Brain. J Neurotrauma. 2006;23(6):950-961. 
doi:10.1089/neu.2006.23.950 

 135 

39.  Osteen CL, Giza CC, Hovda DA. Injury-induced alterations in N-methyl-d-aspartate receptor 
subunit composition contribute to prolonged 45calcium accumulation following lateral fluid 
percussion. Neuroscience. 2004;128(2):305-322. doi:10.1016/j.neuroscience.2004.06.034 

40.  Kandel E, Schwartz J, Jessell T, Siegelbaum S, Hudspeth A, Mack S. Membrane Potential and 

the Passive Electrical Properties of the Neuron. In: Principles of Neuroscience. 6th ed. McGraw HIll. 
Accessed December 1, 2019. 
https://neurology.mhmedical.com/content.aspx?bookid=1049&sectionid=59138628 

41.  Fehily B, Fitzgerald M. Repeated Mild Traumatic Brain Injury. Cell Transplant. 2017;26(7):1131-

1155. doi:10.1177/0963689717714092 

42.  Chen SF, Richards HK, Smielewski P, et al. Relationship Between Flow-Metabolism Uncoupling 
and Evolving Axonal Injury After Experimental Traumatic Brain Injury. Published online 2004. 
doi:10.1097/01.WCB.0000129415.34520.47 

43.  Hill CS, Coleman MP, Menon DK. Traumatic Axonal Injury: Mechanisms and Translational 

Opportunities. Trends Neurosci. 2016;39(5):311-324. doi:10.1016/j.tins.2016.03.002 

44.  Pettus EH, Christman CW, Giebel ML, Povlishock JT. Traumatically Induced Altered 

Membrane Permeability: Its Relationship to Traumatically Induced Reactive Axonal Change. J 
Neurotrauma. 1994;11(5):507-522. doi:10.1089/neu.1994.11.507 

45.  Oakes S. Cell Injury, Cell Death, and Adaptations. In: Robbins & Cotran Pathologic Basis of Disease. 

10th ed. Elsevier. Accessed December 29, 2023. https://www-clinicalkey-
com.proxy1.cl.msu.edu/#!/content/book/3-s2.0-B9780323531139000029 

46.  Staal J, Dickson T, Gasperini R, Liu Y, Foa L, Vickers J. Initial calcium release from intracellular 
stores followed by calcium dysregulation is linked to secondary axotomy following transient 
axonal stretch injury. Accessed December 29, 2023. 
https://onlinelibrary.wiley.com/doi/10.1111/j.1471-4159.2009.06531.x 

47.  Wang Y, Nelson LD, LaRoche AA, et al. Cerebral Blood Flow Alterations in Acute Sport-
Related Concussion. J Neurotrauma. 2016;33(13):1227-1236. doi:10.1089/neu.2015.4072 

48.  Loane DJ, Byrnes KR. Role of microglia in neurotrauma. Neurotherapeutics. 2010;7(4):366-377. 

doi:10.1016/j.nurt.2010.07.002 

49.  Donat CK, Scott G, Gentleman SM, Sastre M. Microglial Activation in Traumatic Brain Injury. 

Front Aging Neurosci. 2017;9:208. doi:10.3389/fnagi.2017.00208 

50.  Csuka E, Hans VHJ, Ammann E, Trentz O, Kossmann T, Morganti-Kossmann MC. Cell 

activation and in¯ammatory response following traumatic axonal injury in the rat. 

51.  Sulhan S, Lyon KA, Shapiro LA, Huang JH. Neuroinflammation and Blood-Brain Barrier 
Disruption Following Traumatic Brain Injury: Pathophysiology and Potential Therapeutic 
Targets. J Neurosci Res. 2020;98(1):19-28. doi:10.1002/jnr.24331 

 136 

52.  Hsieh HL, Wang HH, Wu WB, Chu PJ, Yang CM. Transforming growth factor-b1 induces 

matrix metalloproteinase-9 and cell migration in astrocytes: roles of ROS-dependent ERK- and 
JNK- NF- B pathways. Published online 2010. 

53.  Yang Y, Estrada EY, Thompson JF, Liu W, Rosenberg GA. Matrix Metalloproteinase-Mediated 
Disruption of Tight Junction Proteins in Cerebral Vessels is Reversed by Synthetic Matrix 
Metalloproteinase Inhibitor in Focal Ischemia in Rat. J Cereb Blood Flow Metab. 2007;27(4):697-
709. doi:10.1038/sj.jcbfm.9600375 

54.  Abdul-Muneer PM, Pfister BJ, Haorah J, Chandra N. Role of Matrix Metalloproteinases in the 

Pathogenesis of Traumatic Brain Injury. Mol Neurobiol. 2016;53(9):6106-6123. 
doi:10.1007/s12035-015-9520-8 

55.  Giza C, Greco T, Prins ML. Concussion: pathophysiology and clinical translation. In: Handbook 
of Clinical Neurology. Vol 158. Elsevier; 2018:51-61. doi:10.1016/B978-0-444-63954-7.00006-9 

56.  Strong AJ, Fabricius M, Boutelle MG, et al. Spreading and Synchronous Depressions of Cortical 

Activity in Acutely Injured Human Brain. Stroke. 2002;33(12):2738-2743. 
doi:10.1161/01.STR.0000043073.69602.09 

57.  Mouzon BC, Bachmeier C, Ferro A, et al. Chronic neuropathological and neurobehavioral 
changes in a repetitive mild traumatic brain injury model. Ann Neurol. 2014;75(2):241-254. 
doi:10.1002/ana.24064 

58.  Gsell W, Burke M, Wiedermann D, et al. Differential Effects of NMDA and AMPA Glutamate 
Receptors on Functional Magnetic Resonance Imaging Signals and Evoked Neuronal Activity 
during Forepaw Stimulation of the Rat. J Neurosci. 2006;26(33):8409-8416. 
doi:10.1523/JNEUROSCI.4615-05.2006 

59.  Eckner JT, Kutcher JS, Broglio SP, Richardson JK. Effect of sport-related concussion on 

clinically measured simple reaction time. Br J Sports Med. 2014;48(2):112-118. 
doi:10.1136/bjsports-2012-091579 

60.  Theadom A, Cropley M, Parmar P, et al. Sleep difficulties one year following mild traumatic 

brain injury in a population-based study. Sleep Med. 2015;16(8):926-932. 
doi:10.1016/j.sleep.2015.04.013 

61.  Wickwire EM, Williams SG, Roth T, et al. Sleep, Sleep Disorders, and Mild Traumatic Brain 
Injury. What We Know and What We Need to Know: Findings from a National Working 
Group. Neurotherapeutics. 2016;13(2):403-417. doi:10.1007/s13311-016-0429-3 

62.  Wickwire EM, Schnyer DM, Germain A, et al. Sleep, Sleep Disorders, and Circadian Health 

following Mild Traumatic Brain Injury in Adults: Review and Research Agenda. J Neurotrauma. 
2018;35(22):2615-2631. doi:10.1089/neu.2017.5243 

63.  Sandsmark DK, Elliott JE, Lim MM. Sleep-wake disturbances after traumatic brain injury: 
Synthesis of human and animal studies. Sleep. 2017;40(5). doi:10.1093/sleep/zsx044 

 137 

64.  Popovitz J, Mysore SP, Adwanikar H. Long-Term Effects of Traumatic Brain Injury on 
Anxiety-Like Behaviors in Mice: Behavioral and Neural Correlates. Front Behav Neurosci. 
2019;13(January):1-12. doi:10.3389/fnbeh.2019.00006 

65.  Reger ML, Poulos AM, Buen F, Giza CC, Hovda DA, Fanselow MS. Concussive brain injury 
enhances fear learning and excitatory processes in the amygdala. Biol Psychiatry. 2012;71(4):335-
343. doi:10.1016/j.biopsych.2011.11.007 

66.  Gentleman SM, Leclercq PD, Moyes L, et al. Long-term intracerebral inflammatory response 

after traumatic brain injury. Forensic Sci Int. 2004;146(2-3):97-104. 
doi:10.1016/j.forsciint.2004.06.027 

67.  Figueiredo-Pereira ME, Rockwell P, Schmidt-Glenewinkel T, Serrano P. Neuroinflammation 

and J2 prostaglandins: linking impairment of the ubiquitin-proteasome pathway and 
mitochondria to neurodegeneration. Front Mol Neurosci. 2015;7. doi:10.3389/fnmol.2014.00104 

68.  Tauber D, Parker R. 15-Deoxy-Δ12,14-prostaglandin J2 promotes phosphorylation of 
eukaryotic initiation factor 2α and activates the integrated stress response. J Biol Chem. 
2019;294(16):6344-6352. doi:10.1074/jbc.RA118.007138 

69.  Pakos-Zebrucka K, Koryga I, Mnich K, Ljujic M, Samali A, Gorman AM. The integrated stress 

response. EMBO Rep. 2016;17(10):1374-1395. doi:10.15252/embr.201642195 

70.  Hickey RW, Adelson PD, Johnnides MJ, et al. Cyclooxygenase-2 Activity Following Traumatic 

Brain Injury in the Developing Rat. Pediatr Res. 2007;62(3):271-276. 
doi:10.1203/PDR.0b013e3180db2902 

71.  Kunz T, Marklund N, Hillered L, Oliw EH. Cyclooxygenase-2, Prostaglandin Synthases, and 

Prostaglandin H 2 Metabolism in Traumatic Brain Injury in the Rat. J Neurotrauma. 
2002;19(9):1051-1064. doi:10.1089/089771502760341965 

72.  Chung D eun C, Roemer S, Petrucelli L, Dickson DW. Cellular and pathological heterogeneity 
of primary tauopathies. Mol Neurodegener. 2021;16(1):57. doi:10.1186/s13024-021-00476-x 

73.  Smith DH, Hicks RR, Johnson VE, et al. Pre-Clinical Traumatic Brain Injury Common Data 

Elements: Toward a Common Language Across Laboratories. J Neurotrauma. 2015;32(22):1725-
1735. doi:10.1089/neu.2014.3861 

74.  Wojnarowicz MW, Fisher AM, Minaeva O, Goldstein LE. Considerations for Experimental 

Animal Models of Concussion, Traumatic Brain Injury, and Chronic Traumatic 
Encephalopathy—These Matters Matter. Front Neurol. 2017;8:240. 
doi:10.3389/fneur.2017.00240 

75.  Mullally WJ. Concussion. Am J Med. 2017;130(8):885-892. doi:10.1016/j.amjmed.2017.04.016 

76.  Diaz-Arrastia R, Wang KKW, Papa L, et al. Acute Biomarkers of Traumatic Brain Injury: 

Relationship between Plasma Levels of Ubiquitin C-Terminal Hydrolase-L1 and Glial Fibrillary 
Acidic Protein. J Neurotrauma. 2014;31(1):19-25. doi:10.1089/neu.2013.3040 

 138 

77.  Bacioglu M, Maia LF, Preische O, et al. Neurofilament Light Chain in Blood and CSF as Marker 

of Disease Progression in Mouse Models and in Neurodegenerative Diseases. Neuron. 
2016;91(1):56-66. doi:10.1016/j.neuron.2016.05.018 

78.  Shahim P, Tegner Y, Marklund N, Blennow K, Zetterberg H. Neurofilament light and tau as 

blood biomarkers for sports-related concussion. Neurology. 2018;90(20):e1780-e1788. 
doi:10.1212/WNL.0000000000005518 

79.  Wallace C, Smirl JD, Zetterberg H, et al. Heading in soccer increases serum neurofilament light 

protein and SCAT3 symptom metrics. BMJ Open Sport Exerc Med. 2018;4(1):e000433. 
doi:10.1136/bmjsem-2018-000433 

80.  Wirsching A, Chen Z, Bevilacqua ZW, Huibregtse ME, Kawata K. Association of Acute 

Increase in Plasma Neurofilament Light with Repetitive Subconcussive Head Impacts: A Pilot 
Randomized Control Trial. J Neurotrauma. 2019;36(4):548-553. doi:10.1089/neu.2018.5836 

81.  Rubin LH, Tierney R, Kawata K, et al. NFL blood levels are moderated by subconcussive 

impacts in a cohort of college football players. Brain Inj. 2019;33(4):456-462. 
doi:10.1080/02699052.2019.1565895 

82.  Korley FK, Datwyler SA, Jain S, et al. Comparison of GFAP and UCH-L1 Measurements from 

Two Prototype Assays: The Abbott i-STAT and ARCHITECT Assays. Neurotrauma Rep. 
2021;2(1):193-199. doi:10.1089/neur.2020.0037 

83.  Manley G. TRACK-TBI: Study Design, Early Deliverables, and Future Directions with 

Geoffrey Manley, MD, PhD. Presented at: 2020; University of Michigan Concussion Center. 
https://www.youtube.com/watch?v=bZEhRhE_PqM 

84.  Patricios JS, Schneider KJ, Dvorak J, et al. Consensus statement on concussion in sport: the 6th 
International Conference on Concussion in Sport–Amsterdam, October 2022. Br J Sports Med. 
2023;57(11):695-711. doi:10.1136/bjsports-2023-106898 

85.  Diffusion MRI from Quantitative Measurement to In-Vivo Neuroanatomy. 2nd ed. Elsevier/Academic 

Press,; 2014. 

86.  Introduction to Diffusion Tensor Imaging and Higher Order Models. 2nd edition /. Elsevier/Academic 

Press,; 2014. 

87.  Cooley JW, Tukey JW. An Algorithm for the Machine Calculation of Complex Fourier Series. 

88.  Bammer R. Basic principles of diffusion-weighted imaging. Eur J Radiol. 2003;45(3):169-184. 

doi:10.1016/S0720-048X(02)00303-0 

89.  Pipe J. Pulse Sequences for Diffusion-Weighted MRI. In: Diffusion MRI. Elsevier; 2014:11-34. 

doi:10.1016/B978-0-12-396460-1.00002-0 

90.  Novikov DS, Fieremans E, Jespersen SN, Kiselev VG. Quantifying brain microstructure with 

diffusion MRI: Theory and parameter estimation. NMR Biomed. 2019;32(4):e3998. 
doi:10.1002/nbm.3998 

 139 

91.  Basser PJ, Özarslan E. Introduction to Diffusion MR. In: Diffusion MRI. Elsevier; 2014:3-9. 

doi:10.1016/B978-0-12-396460-1.00001-9 

92.  Einstein A. Investigations on the Theory of the Brownian Movement. 

93.  Anatomy of Diffusion Measurement. In: Introduction to Diffusion Tensor Imaging. Elsevier; 2014:11-

15. doi:10.1016/B978-0-12-398398-5.00002-3 

94.  Stejskal EO, Tanner JE. Spin Diffusion Measurements: Spin Echoes in the Presence of a Time-

Dependent Field Gradient. J Chem Phys. 1965;42(1):288-292. doi:10.1063/1.1695690 

95.  Principle of Diffusion Tensor Imaging. In: Introduction to Diffusion Tensor Imaging. Elsevier; 

2014:27-32. doi:10.1016/B978-0-12-398398-5.00004-7 

96.  Learning Radiology - 9780323878173 | Elsevier Health. MEA Elsevier Health. Accessed March 

3, 2024. https://www.mea.elsevierhealth.com/learning-radiology-9780323878173.html 

97.  Bastin ME, Le Roux P. On the application of a non-CPMG single-shot fast spin-echo sequence 

to diffusion tensor MRI of the human brain. Magn Reson Med. 2002;48(1):6-14. 
doi:10.1002/mrm.10214 

98.  Elster AD. Gradient-Echo MR Imaging: Techniques and Acronyms. Radiology. 1993;186(1):1-8. 

doi:10.1148/radiology.186.1.8416546. 

99.  Xu D, Henry RG, Mukherjee P, et al. Single-shot fast spin-echo diffusion tensor imaging of the 

brain and spine with head and phased array coils at 1.5 T and 3.0 T. Magn Reson Imaging. 
2004;22(6):751-759. doi:10.1016/j.mri.2004.01.075 

100. Basser PJ, Mattiello J, LeBihan D. MR diffusion tensor spectroscopy and imaging. Biophys J. 

1994;66(1):259-267. doi:10.1016/S0006-3495(94)80775-1 

101. Basser PJ, Mattiello J, LeBihan D. Estimation of the Effective Self-Diffusion Tensor from the 
NMR Spin Echo. J Magn Reson B. 1994;103(4):247-254. doi:doi: 10.1006/jmrb.1994.1037. 

102. Basser PJ, LeBihant D. Fiber orientation mapping in an anisotropic medium with NMR 

diffusion spectroscopy. 

103. Basser PJ, Zierler D. Oral History Interview with Peter Basser. 

104. Moving Beyond DTI. In: Introduction to Diffusion Tensor Imaging. Elsevier; 2014:65-78. 

doi:10.1016/B978-0-12-398398-5.00008-4 

105. Jones DK, Knösche TR, Turner R. White matter integrity, fiber count, and other fallacies: the 

do’s and don’ts of diffusion MRI. NeuroImage. 2013;73:239-254. 
doi:10.1016/j.neuroimage.2012.06.081 

106. Mathematics of Diffusion Tensor Imaging. In: Introduction to Diffusion Tensor Imaging. Elsevier; 

2014:33-37. doi:10.1016/B978-0-12-398398-5.00005-9 

 140 

107. Beauchamp N, Ulug A, Passe T, van Zijl P. MR Diffusion Imaging in Stroke: Review and 

Controversies. Radiographics. 1998;18:1269-1283. 

108. Apparent Diffusion. Questions and Answers in MRI. Accessed March 3, 2024. 

http://mriquestions.com/apparent-diffusion.html 

109. DTI. Questions and Answers in MRI. Accessed March 3, 2024. http://mriquestions.com/dti-

tensor-imaging.html 

110. Chary K, Manninen E, Claessens J, Ramirez-Manzanares A, Gröhn O, Sierra A. Diffusion MRI 
approaches for investigating microstructural complexity in a rat model of traumatic brain injury. 
Sci Rep. 2023;13(1):2219. doi:10.1038/s41598-023-29010-3 

111. Chary K, Narvaez O, Salo RA, et al. Microstructural Tissue Changes in a Rat Model of Mild 
Traumatic Brain Injury. Front Neurosci. 2021;15:746214. doi:10.3389/fnins.2021.746214 

112. TRACK-TBI Investigators. Transforming Research and Clinical Knowledge in Traumatic Brain 

Injury Clinical Protocol. Published online 18 2018. Accessed June 16, 2023. 
https://tracktbi.ucsf.edu/sites/tracktbi.ucsf.edu/files/TRACKTBI%20U01%20Clinical%20Pro
tocol%20V18-January%2018%202019.pdf 

113. TRACK-TBI Precision Medicine Neuroimaging MOP v1.1. Published online March 10, 2021. 

114. Oishi K, ed. MRI Atlas of Human White Matter. 2. ed. Academic Press; 2011. 

115. Palacios EM, Yuh EL, Mac Donald CL, et al. Diffusion Tensor Imaging Reveals Elevated 

Diffusivity of White Matter Microstructure that Is Independently Associated with Long-Term 
Outcome after Mild Traumatic Brain Injury: A TRACK-TBI Study. J Neurotrauma. 2022;39(19-
20):1318-1328. doi:10.1089/neu.2021.0408 

116. Zhang B, Daneshvar DH, Polich G, Glenn MB. Response to Machamer et al., “Symptom 
Frequency and Persistence in the First Year after Traumatic Brain Injury: A TRACK-TBI 
Study” (doi: 10.1089/neu.2021.0348). J Neurotrauma. 2023;40(5-6):595-596. 
doi:10.1089/neu.2022.0292 

117. Chiang CW, Wang Y, Sun P, et al. Quantifying white matter tract diffusion parameters in the 

presence of increased extra-fiber cellularity and vasogenic edema. NeuroImage. 2014;101:310-319. 
doi:10.1016/j.neuroimage.2014.06.064 

118. Chu Z, Wilde EA, Hunter JV, et al. Voxel-Based Analysis of Diffusion Tensor Imaging in Mild 

Traumatic Brain Injury in Adolescents. Am J Neuroradiol. 2010;31(2):340-346. 
doi:10.3174/ajnr.A1806 

119. Wilde EA, McCauley SR, Hunter JV, et al. Diffusion tensor imaging of acute mild traumatic 

brain injury in adolescents. Neurology. 2008;70(12):948-955. 
doi:10.1212/01.wnl.0000305961.68029.54 

 141 

120. Palacios EM, Owen JP, Yuh EL, et al. The evolution of white matter microstructural changes 

after mild traumatic brain injury: A longitudinal DTI and NODDI study. Sci Adv. 
2020;6(32):eaaz6892. doi:10.1126/sciadv.aaz6892 

121. Raffelt D, Crozier S, Connelly A, Salvado O, Tournier JD. Apparent Fibre Density: A New 

Measure for High Angular Resolution Diffusion-Weighted Image Analysis. 

122. Wu YC, Mustafi SM, Harezlak J, Kodiweera C, Flashman LA, McAllister TW. Hybrid 

Diffusion Imaging in Mild Traumatic Brain Injury. J Neurotrauma. 2018;35(20):2377-2390. 
doi:10.1089/neu.2017.5566 

123. Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, Cross AH. Dysmyelination Revealed 
through MRI as Increased Radial (but Unchanged Axial) Diffusion of Water. NeuroImage. 
2002;17(3):1429-1436. doi:10.1006/nimg.2002.1267 

124. Song SK, Yoshino J, Le TQ, et al. Demyelination increases radial diffusivity in corpus callosum 

of mouse brain. NeuroImage. 2005;26(1):132-140. doi:10.1016/j.neuroimage.2005.01.028 

125. Zhu DC, Covassin T, Nogle S, et al. A Potential Biomarker in Sports-Related Concussion: 
Brain Functional Connectivity Alteration of the Default-Mode Network Measured with 
Longitudinal Resting-State fMRI over Thirty Days. J Neurotrauma. 2015;32(5):327-341. 
doi:10.1089/neu.2014.3413 

126. Raffelt DA, Smith RE, Ridgway GR, et al. Connectivity-based fixel enhancement: Whole-brain 
statistical analysis of diffusion MRI measures in the presence of crossing fibres. NeuroImage. 
2015;117:40-55. doi:10.1016/j.neuroimage.2015.05.039 

127. Winston GP. The physical and biological basis of quantitative parameters derived from 

diffusion MRI. Quant Imaging Med Surg. 2012;2(4). doi:10.3978/j.issn.2223-4292.2012.12.05 

128. Jensen JH, Helpern JA, Ramani A, Lu H, Kaczynski K. Diffusional kurtosis imaging: The 

quantification of non-gaussian water diffusion by means of magnetic resonance imaging. Magn 
Reson Med. 2005;53(6):1432-1440. doi:10.1002/mrm.20508 

129. Hansen B, Lund TE, Sangill R, Jespersen SN. Experimentally and computationally fast method 

for estimation of a mean kurtosis. Magn Reson Med. 2013;69(6):1754-1760. 
doi:10.1002/mrm.24743 

130. Yan X, Zhou M, Ying L, et al. Evaluation of optimized b-value sampling schemas for diffusion 

kurtosis imaging with an application to stroke patient data. Comput Med Imaging Graph. 
2013;37(4):272-280. doi:10.1016/j.compmedimag.2013.04.007 

131. Steven AJ, Zhuo J, Melhem ER. Diffusion Kurtosis Imaging: An Emerging Technique for 
Evaluating the Microstructural Environment of the Brain. Am J Roentgenol. 2014;202(1):W26-
W33. doi:10.2214/AJR.13.11365 

132. Zhuo J, Xu S, Proctor JL, et al. Diffusion kurtosis as an in vivo imaging marker for reactive 

astrogliosis in traumatic brain injury. NeuroImage. 2012;59(1):467-477. 
doi:10.1016/j.neuroimage.2011.07.050 

 142 

133. Jespersen SN, Kroenke CD, Østergaard L, Ackerman JJH, Yablonskiy DA. Modeling dendrite 
density from magnetic resonance diffusion measurements. NeuroImage. 2007;34(4):1473-1486. 
doi:10.1016/j.neuroimage.2006.10.037 

134. Zhang H, Schneider T, Wheeler-Kingshott CA, Alexander DC. NODDI: Practical in vivo 

neurite orientation dispersion and density imaging of the human brain. NeuroImage. 
2012;61(4):1000-1016. doi:10.1016/j.neuroimage.2012.03.072 

135. Zhang H, Hubbard PL, Parker GJM, Alexander DC. Axon diameter mapping in the presence 

of orientation dispersion with diffusion MRI. NeuroImage. 2011;56(3):1301-1315. 
doi:10.1016/j.neuroimage.2011.01.084 

136. Tournier JD, Calamante F, Connelly A. Robust determination of the fibre orientation 
distribution in diffusion MRI: Non-negativity constrained super-resolved spherical 
deconvolution. NeuroImage. 2007;35(4):1459-1472. doi:10.1016/j.neuroimage.2007.02.016 

137. Raffelt D, Tournier JD, Fripp J, Crozier S, Connelly A, Salvado O. Symmetric diffeomorphic 

registration of fibre orientation distributions. NeuroImage. 2011;56(3):1171-1180. 
doi:10.1016/j.neuroimage.2011.02.014 

138. Smith SM, Jenkinson M, Johansen-Berg H, et al. Tract-based spatial statistics: Voxelwise 

analysis of multi-subject diffusion data. NeuroImage. 2006;31(4):1487-1505. 
doi:10.1016/j.neuroimage.2006.02.024 

139. Tournier JD, Calamante F, Connelly A. Improved probabilistic streamlines tractography by 2nd 

order integration over fibre orientation distributions. 

140. Smith S, Nichols T. Threshold-free cluster enhancement: Addressing problems of smoothing, 
threshold dependence and localisation in cluster inference. NeuroImage. 2009;44(1):83-98. 
doi:10.1016/j.neuroimage.2008.03.061 

141. Glenn GR, Kuo LW, Chao YP, Lee CY, Helpern JA, Jensen JH. Mapping the Orientation of 
White Matter Fiber Bundles: A Comparative Study of Diffusion Tensor Imaging, Diffusional 
Kurtosis Imaging, and Diffusion Spectrum Imaging. Am J Neuroradiol. 2016;37(7):1216-1222. 
doi:10.3174/ajnr.A4714 

142. Grossman EJ, Ge Y, Jensen JH, et al. Thalamus and Cognitive Impairment in Mild Traumatic 
Brain Injury: A Diffusional Kurtosis Imaging Study. J Neurotrauma. 2012;29(13):2318-2327. 
doi:10.1089/neu.2011.1763 

143. Stenberg J, Eikenes L, Moen KG, Vik A, Håberg AK, Skandsen T. Acute Diffusion Tensor and 

Kurtosis Imaging and Outcome following Mild Traumatic Brain Injury. J Neurotrauma. 
2021;38(18):2560-2571. doi:10.1089/neu.2021.0074 

144. Stenberg J, Skandsen T, Gøran Moen K, Vik A, Eikenes L, Håberg AK. Diffusion Tensor and 
Kurtosis Imaging Findings the First Year following Mild Traumatic Brain Injury. J Neurotrauma. 
2023;40(5-6):457-471. doi:10.1089/neu.2022.0206 

 143 

145. Lancaster MA, Meier TB, Olson DV, McCrea MA, Nelson LD, Muftuler LT. Chronic 

differences in white matter integrity following sport-related concussion as measured by diffusion 
MRI: 6-Month follow-up. Hum Brain Mapp. 2018;39(11):4276-4289. doi:10.1002/hbm.24245 

146. Lancaster MA, Olson DV, McCrea MA, Nelson LD, LaRoche AA, Muftuler LT. Acute white 
matter changes following sport-related concussion: A serial diffusion tensor and diffusion 
kurtosis tensor imaging study. Hum Brain Mapp. 2016;37(11):3821-3834. doi:10.1002/hbm.23278 

147. Muftuler LT, Meier TB, Keith M, Budde MD, Huber DL, McCrea MA. Serial Diffusion 

Kurtosis Magnetic Resonance Imaging Study during Acute, Subacute, and Recovery Periods 
after Sport-Related Concussion. J Neurotrauma. 2020;37(19):2081-2092. 
doi:10.1089/neu.2020.6993 

148. Chung S, Chen J, Li T, Wang Y, Lui YW. Investigating Brain White Matter in Football Players 
with and without Concussion Using a Biophysical Model from Multishell Diffusion MRI. Am J 
Neuroradiol. 2022;43(6):823-828. doi:10.3174/ajnr.A7522 

149. Brett BL, Koch KM, Muftuler LT, Budde M, McCrea MA, Meier TB. Association of Head 

Impact Exposure with White Matter Macrostructure and Microstructure Metrics. J Neurotrauma. 
2021;38(4):474-484. doi:10.1089/neu.2020.7376 

150. Stokum JA, Sours C, Zhuo J, Kane R, Shanmuganathan K, Gullapalli RP. A longitudinal 

evaluation of diffusion kurtosis imaging in patients with mild traumatic brain injury. Brain Inj. 
2015;29(1):47-57. doi:10.3109/02699052.2014.947628 

151. Wang ML, Wei XE, Yu MM, Li WB. Cognitive impairment in mild traumatic brain injury: a 

diffusion kurtosis imaging and volumetric study. Acta Radiol. 2022;63(4):504-512. 
doi:10.1177/0284185121998317 

152. Næss-Schmidt ET, Blicher JU, Eskildsen SF, et al. Microstructural changes in the thalamus 

after mild traumatic brain injury: A longitudinal diffusion and mean kurtosis tensor MRI study. 
Brain Inj. 2017;31(2):230-236. doi:10.1080/02699052.2016.1229034 

153. Grossman EJ, Jensen JH, Babb JS, et al. Cognitive Impairment in Mild Traumatic Brain Injury: 
A Longitudinal Diffusional Kurtosis and Perfusion Imaging Study. AJNR Am J Neuroradiol. 
2013;34(5):951-957. doi:10.3174/ajnr.A3358 

154. Gong NJ, Kuzminski S, Clark M, et al. Microstructural alterations of cortical and deep gray 
matter over a season of high school football revealed by diffusion kurtosis imaging. Neurobiol 
Dis. 2018;119:79-87. doi:10.1016/j.nbd.2018.07.020 

155. Davenport EM, Apkarian K, Whitlow CT, et al. Abnormalities in Diffusional Kurtosis Metrics 

Related to Head Impact Exposure in a Season of High School Varsity Football. J Neurotrauma. 
2016;33(23):2133-2146. doi:10.1089/neu.2015.4267 

156. Goubran M, Mills BD, Georgiadis M, et al. Microstructural Alterations in Tract Development 
in College Football and Volleyball Players: A Longitudinal Diffusion MRI Study. Neurology. 
2023;101(9). doi:10.1212/WNL.0000000000207543 

 144 

157. Karlsen RH, Einarsen C, Moe HK, et al. Diffusion kurtosis imaging in mild traumatic brain 

injury and postconcussional syndrome. J Neurosci Res. 2019;97(5):568-581. doi:10.1002/jnr.24383 

158. Churchill NW, Caverzasi E, Graham SJ, Hutchison MG, Schweizer TA. White matter during 
concussion recovery: Comparing diffusion tensor imaging (DTI) and neurite orientation 
dispersion and density imaging (NODDI). Hum Brain Mapp. 2019;40(6):1908-1918. 
doi:10.1002/hbm.24500 

159. Mayer AR, Ling JM, Dodd AB, Meier TB, Hanlon FM, Klimaj SD. A prospective 

microstructure imaging study in mixed-martial artists using geometric measures and diffusion 
tensor imaging: methods and findings. Brain Imaging Behav. 2017;11(3):698-711. 
doi:10.1007/s11682-016-9546-1 

160. Huang S, Huang C, Li M, Zhang H, Liu J. White Matter Abnormalities and Cognitive Deficit 

After Mild Traumatic Brain Injury: Comparing DTI, DKI, and NODDI. Front Neurol. 
2022;13:803066. doi:10.3389/fneur.2022.803066 

161. Oehr LE, Yang JYM, Chen J, Maller JJ, Seal ML, Anderson JFI. Investigating White Matter 

Tract Microstructural Changes at Six–Twelve Weeks following Mild Traumatic Brain Injury: A 
Combined Diffusion Tensor Imaging and Neurite Orientation Dispersion and Density Imaging 
Study. J Neurotrauma. 2021;38(16):2255-2263. doi:10.1089/neu.2020.7310 

162. Shukla A, Ware AL, Guo S, et al. Examining brain white matter after pediatric mild traumatic 

brain injury using neurite orientation dispersion and density imaging: An A-CAP study. 
NeuroImage Clin. 2021;32:102887. doi:10.1016/j.nicl.2021.102887 

163. Seider NA, Adeyemo B, Miller R, et al. Accuracy and reliability of diffusion imaging models. 

NeuroImage. 2022;254:119138. doi:10.1016/j.neuroimage.2022.119138 

164. Mayer AR, Ling JM, Dodd AB, et al. Multicompartmental models and diffusion abnormalities 

in paediatric mild traumatic brain injury. Brain. 2022;145(11):4124-4137. 
doi:10.1093/brain/awac221 

165. Harms RL, Fritz FJ, Tobisch A, Goebel R, Roebroeck A. Robust and fast nonlinear 
optimization of diffusion MRI microstructure models. NeuroImage. 2017;155:82-96. 
doi:10.1016/j.neuroimage.2017.04.064 

166. Stein A, Vinh To X, Nasrallah FA, Barlow KM. Evidence of Ongoing Cerebral Microstructural 

Reorganization in Children With Persisting Symptoms Following Mild Traumatic Brain Injury: 
A NODDI DTI Analysis. J Neurotrauma. 2024;41(1-2):41-58. doi:10.1089/neu.2023.0196 

167. Lima Santos JP, Kontos AP, Holland CL, et al. The role of sleep quality on white matter 

integrity and concussion symptom severity in adolescents. NeuroImage Clin. 2022;35:103130. 
doi:10.1016/j.nicl.2022.103130 

168. Muller J, Middleton D, Alizadeh M, et al. Hybrid diffusion imaging reveals altered white matter 
tract integrity and associations with symptoms and cognitive dysfunction in chronic traumatic 
brain injury. NeuroImage Clin. 2021;30:102681. doi:10.1016/j.nicl.2021.102681 

 145 

169. Anderson JFI, Oehr LE, Chen J, Maller JJ, Seal ML, Yang JYM. The relationship between 
cognition and white matter tract damage after mild traumatic brain injury in a premorbidly 
healthy, hospitalised adult cohort during the post-acute period. Front Neurol. 2023;14:1278908. 
doi:10.3389/fneur.2023.1278908 

170. Kawata K, Steinfeldt JA, Huibregtse ME, et al. Association Between Proteomic Blood 

Biomarkers and DTI/NODDI Metrics in Adolescent Football Players: A Pilot Study. Front 
Neurol. 2020;11:581781. doi:10.3389/fneur.2020.581781 

171. Farquharson S, Tournier JD, Calamante F, et al. White matter fiber tractography: why we need 

to move beyond DTI: Clinical article. J Neurosurg. 2013;118(6):1367-1377. 
doi:10.3171/2013.2.JNS121294 

172. Mohammadian M, Roine T, Hirvonen J, et al. High angular resolution diffusion-weighted 

imaging in mild traumatic brain injury. NeuroImage Clin. 2016;13:174-180. 
doi:10.1016/j.nicl.2016.11.016 

173. Van Der Horn HJ, Kok JG, De Koning ME, et al. Altered Wiring of the Human Structural 

Connectome in Adults with Mild Traumatic Brain Injury. J Neurotrauma. 2017;34(5):1035-1044. 
doi:10.1089/neu.2016.4659 

174. Wallace EJ, Mathias JL, Ward L, Fripp J, Rose S, Pannek K. A fixel-based analysis of micro- 

and macro-structural changes to white matter following adult traumatic brain injury. Hum Brain 
Mapp. 2020;41(8):2187-2197. doi:10.1002/hbm.24939 

175. Winston GP. The physical and biological basis of quantitative parameters derived from 
diffusion MRI. Quant Imaging Med Surg. 2012;2(4):254-265. doi:10.1093/cercor/bhab095 

176. Mito R, Parker DM, Abbott DF, Makdissi M, Pedersen M, Jackson GD. White matter 

abnormalities characterize the acute stage of sports-related mild traumatic brain injury. Brain 
Commun. 2022;4(4):fcac208. doi:10.1093/braincomms/fcac208 

177. Bach M, Laun FB, Leemans A, et al. Methodological considerations on tract-based spatial 
statistics (TBSS). NeuroImage. 2014;100:358-369. doi:10.1016/j.neuroimage.2014.06.021 

178. Kelley S, Plass J, Bender AR, Polk TA. Age-Related Differences in White Matter: 

Understanding Tensor-Based Results Using Fixel-Based Analysis. Cereb Cortex N Y NY. 
2021;31(8):3881-3898. doi:10.1093/cercor/bhab056 

179. Daducci A, Canales-Rodríguez EJ, Zhang H, Dyrby TB, Alexander DC, Thiran JP. Accelerated 

Microstructure Imaging via Convex Optimization (AMICO) from diffusion MRI data. 
NeuroImage. 2015;105:32-44. doi:10.1016/j.neuroimage.2014.10.026 

180. Faiyaz A, Doyley M, Schifitto G, Zhong J, Uddin MN. Single-shell NODDI using dictionary-

learner-estimated isotropic volume fraction. NMR Biomed. 2022;35(2):e4628. 
doi:10.1002/nbm.4628 

181. Mikula S, Binding J, Denk W. Staining and embedding the whole mouse brain for electron 

microscopy. Nat Methods. 2012;9(12):1198-1201. doi:10.1038/nmeth.2213 

 146 

182. Jeurissen B, Leemans A, Tournier J, Jones DK, Sijbers J. Investigating the prevalence of 

complex fiber configurations in white matter tissue with diffusion magnetic resonance imaging. 
Hum Brain Mapp. 2013;34(11):2747-2766. doi:10.1002/hbm.22099 

183. Schilling KG, Tax CMW, Rheault F, et al. Prevalence of white matter pathways coming into a 
single white matter voxel orientation: The bottleneck issue in tractography. Hum Brain Mapp. 
2022;43(4):1196-1213. doi:10.1002/hbm.25697 

184. Gazdzinski LM, Mellerup M, Wang T, et al. White Matter Changes Caused by Mild Traumatic 

Brain Injury in Mice Evaluated Using Neurite Orientation Dispersion and Density Imaging. J 
Neurotrauma. 2020;37(16):1818-1828. doi:10.1089/neu.2020.6992 

185. McCunn P, Xu X, Moszczynski A, Li A, Brown A, Bartha R. Neurite orientation dispersion and 

density imaging in a rodent model of acute mild traumatic brain injury. J Neuroimaging. 
2021;31(5):879-892. doi:10.1111/jon.12917 

186. Valera EM, Joseph ALC, Snedaker K, et al. Understanding Traumatic Brain Injury in Females: 
A State-of-the-Art Summary and Future Directions. J Head Trauma Rehabil. 2021;36(1):E1-E17. 
doi:10.1097/HTR.0000000000000652 

187. Levin HS, Temkin NR, Barber J, et al. Association of Sex and Age With Mild Traumatic Brain 
Injury–Related Symptoms: A TRACK-TBI Study. JAMA Netw Open. 2021;4(4):e213046. 
doi:10.1001/jamanetworkopen.2021.3046 

188. Wickwire EM, Albrecht JS, Capaldi VF, et al. Trajectories of Insomnia in Adults After 

Traumatic Brain Injury. JAMA Netw Open. 2022;5(1):e2145310. 
doi:10.1001/jamanetworkopen.2021.45310 

189. Næss-Schmidt ET, Blicher JU, Tietze A, et al. Diffusion MRI findings in patients with 

extensive and minimal post-concussion symptoms after mTBI and healthy controls: a cross 
sectional study. Brain Inj. 2018;32(1):91-98. doi:10.1080/02699052.2017.1377352 

190. Tallus J, Mohammadian M, Kurki T, Roine T, Posti JP, Tenovuo O. A comparison of diffusion 

tensor imaging tractography and constrained spherical deconvolution with automatic 
segmentation in traumatic brain injury. NeuroImage Clin. 2023;37:103284. 
doi:10.1016/j.nicl.2022.103284 

191. Roine T, Mohammadian M, Hirvonen J, et al. Structural Brain Connectivity Correlates with 

Outcome in Mild Traumatic Brain Injury. J Neurotrauma. 2022;39(5-6):336-347. 
doi:10.1089/neu.2021.0093 

192. Thompson HJ, Vavilala MS, Rivara FP. Common Data Elements and Federal Interagency 
Traumatic Brain Injury Research Informatics System for TBI Research. Annu Rev Nurs Res. 
2015;33(1):1-11. doi:10.1891/0739-6686.33.1 

193. Jeurissen B, Tournier JD, Dhollander T, Connelly A, Sijbers J. Multi-tissue constrained 

spherical deconvolution for improved analysis of multi-shell diffusion MRI data. NeuroImage. 
2014;103:411-426. doi:10.1016/j.neuroimage.2014.07.061 

 147 

194. Genc S, Tax CMW, Raven EP, Chamberland M, Parker GD, Jones DK. Impact of b-value on 

estimates of apparent fibre density. Hum Brain Mapp. 2020;41(10):2583-2595. 
doi:10.1002/hbm.24964 

195. Calamuneri A, Arrigo A, Mormina E, et al. White Matter Tissue Quantification at Low b-

Values Within Constrained Spherical Deconvolution Framework. Front Neurol. 2018;9. Accessed 
November 16, 2023. https://www.frontiersin.org/articles/10.3389/fneur.2018.00716 

196. Newman BT, Dhollander T, Reynier KA, Panzer MB, Druzgal TJ. Test-retest reliability and 
long-term stability of 3-tissue constrained spherical deconvolution methods for analyzing 
diffusion MRI data. Magn Reson Med. 2020;84(4):2161-2173. doi:10.1002/mrm.28242 

197. Genc S, Smith RE, Malpas CB, et al. Development of white matter fibre density and 

morphology over childhood: A longitudinal fixel-based analysis. NeuroImage. 2018;183:666-676. 
doi:10.1016/j.neuroimage.2018.08.043 

198. Schilling KG, Blaber J, Huo Y, et al. Synthesized b0 for diffusion distortion correction (Synb0-

DisCo). Magn Reson Imaging. 2019;64:62-70. doi:10.1016/j.mri.2019.05.008 

199. Dhollander T, Mito R, Raffelt D, Connelly A. Improved white matter response function 
estimation for 3-tissue constrained spherical deconvolution. Published online 2019. 

200. Dhollander T, Raffelt D, Connelly A. Unsupervised 3-tissue response function estimation from 
single-shell or multi-shell diffusion MR data without a co-registered T1 image. Published online 
2016. 

201. Raffelt D, Tournier JD, Fripp J, Crozier S, Connelly A, Salvado O. Symmetric diffeomorphic 
registration of fibre orientation distributions. Neuroimage. 2011 Jun 1;56(3):1171-80. doi: 
10.1016/j.neuroimage.2011.02.014. Epub 2011 Feb 18. PMID: 21316463.  

202. Raffelt D, Tournier J, Crozier S, Connelly A, Salvado O. Reorientation of fiber orientation 
distributions using apodized point spread functions. Magn Reson Med. 2012;67(3):844-855. 
doi:10.1002/mrm.23058 

203. Raffelt D, Tournier JD, Rose S, et al. Apparent Fibre Density: A novel measure for the analysis 

of diffusion-weighted magnetic resonance images. NeuroImage. 2012;59(4):3976-3994. 
doi:10.1016/j.neuroimage.2011.10.045 

204. Raffelt DA, Tournier JD, Smith RE, et al. Investigating white matter fibre density and 

morphology using fixel-based analysis. NeuroImage. 2017;144:58-73. 
doi:10.1016/j.neuroimage.2016.09.029 

205. Gardner RC, Yaffe K. Epidemiology of mild traumatic brain injury and neurodegenerative 

disease. Mol Cell Neurosci. 2015;66:75-80. doi:10.1016/j.mcn.2015.03.001 

206. Rao DP, McFaull S, Thompson W, Jayaraman GC. Trends in self-reported traumatic brain 

injury among Canadians, 2005-2014: a repeated cross-sectional analysis. CMAJ Open. 
2017;5(2):E301-E307. doi:10.9778/cmajo.20160115 

 148 

207. Panwar J, Hsu CCT, Tator CH, Mikulis D. Magnetic Resonance Imaging Criteria for Post-
Concussion Syndrome: A Study of 127 Post-Concussion Syndrome Patients. J Neurotrauma. 
2020;37(10):1190-1196. doi:10.1089/neu.2019.6809 

208. Isokuortti H, Iverson GL, Silverberg ND, et al. Characterizing the type and location of 

intracranial abnormalities in mild traumatic brain injury. J Neurosurg. 2018;129(6):1588-1597. 
doi:10.3171/2017.7.JNS17615 

209. Huang W, Hu W, Zhang P, et al. Early Changes in the White Matter Microstructure and 

Connectome Underlie Cognitive Deficit and Depression Symptoms After Mild Traumatic Brain 
Injury. Front Neurol. 2022;13:880902. doi:10.3389/fneur.2022.880902 

210. Dean PJA, Sato JR, Vieira G, McNamara A, Sterr A. Long-term structural changes after mTBI 

and their relation to post-concussion symptoms. Brain Inj. 2015;29(10):1211-1218. 
doi:10.3109/02699052.2015.1035334 

211. Boonstra FM, Clough M, Strik M, et al. Longitudinal tracking of axonal loss using diffusion 

magnetic resonance imaging in multiple sclerosis. Brain Commun. 2022;4(2):fcac065. 
doi:10.1093/braincomms/fcac065 

212. Petersen M, Frey BM, Mayer C, et al. Fixel based analysis of white matter alterations in early 
stage cerebral small vessel disease. Sci Rep. 2022;12(1):1581. doi:10.1038/s41598-022-05665-2 

213. Meisler SL, Gabrieli JD. Fiber-specific structural properties relate to reading skills in children 

and adolescents. eLife. 2022;11:e82088. doi:10.7554/eLife.82088 

214. Wright DK, Symons GF, O’Brien WT, et al. Diffusion Imaging Reveals Sex Differences in the 
White Matter Following Sports-Related Concussion. Cereb Cortex. 2021;31(10):4411-4419. 
doi:10.1093/cercor/bhab095 

215. Wallace EJ, Mathias JL, Ward L, Fripp J, Rose S, Pannek K. A fixel-based analysis of micro- 

and macro-structural changes to white matter following adult traumatic brain injury. Hum Brain 
Mapp. 2020;41(8):2187-2197. doi:10.1002/hbm.24939 

216. Wasserthal J, Neher P, Maier-Hein KH. TractSeg - Fast and accurate white matter tract 
segmentation. NeuroImage. 2018;183:239-253. doi:10.1016/j.neuroimage.2018.07.070 

217. Rohlfing T. Incorrect ICBM-DTI-81 atlas orientation and white matter labels. Front Neurosci. 

2013;7. doi:10.3389/fnins.2013.00004 

218. Chang YS, Owen JP, Pojman NJ, et al. White Matter Changes of Neurite Density and Fiber 

Orientation Dispersion during Human Brain Maturation. Gong G, ed. PLOS ONE. 
2015;10(6):e0123656. doi:10.1371/journal.pone.0123656 

219. Henschel L, Conjeti S, Estrada S, Diers K, Fischl B, Reuter M. FastSurfer - A fast and accurate 

deep learning based neuroimaging pipeline. NeuroImage. 2020;219:117012. 
doi:10.1016/j.neuroimage.2020.117012 

 149 

220. Behrens TEJ, Woolrich MW, Jenkinson M, et al. Characterization and propagation of 
uncertainty in diffusion-weighted MR imaging. Magn Reson Med. 2003;50(5):1077-1088. 
doi:10.1002/mrm.10609 

221. Behrens TEJ, Berg HJ, Jbabdi S, Rushworth MFS, Woolrich MW. Probabilistic diffusion 

tractography with multiple fibre orientations: What can we gain? NeuroImage. 2007;34(1):144-155. 
doi:10.1016/j.neuroimage.2006.09.018 

222. Jbabdi S, Sotiropoulos SN, Savio AM, Graña M, Behrens TEJ. Model-based analysis of 

multishell diffusion MR data for tractography: How to get over fitting problems. Magn Reson 
Med. 2012;68(6):1846-1855. doi:10.1002/mrm.24204 

223. Smith RE, Tournier JD, Calamante F, Connelly A. SIFT: Spherical-deconvolution informed 
filtering of tractograms. NeuroImage. 2013;67:298-312. doi:10.1016/j.neuroimage.2012.11.049 

224. Smith R, Christiaens D, Jeurissen B, et al. On false positive control in Fixel-Based Analysis. Proc 

Intl Soc Mag Reson Med. Published online 2021. 

225. Smith RE, Dhollander T, Connelly A. On the regression of intracranial volume in Fixel-Based 

Analysis. 

226. MATLAB 2022b, The MathWorks, Inc., Natick, Massachusetts, United States. 

227. Storey JD. A Direct Approach to False Discovery Rates. J R Stat Soc Ser B Stat Methodol. 

2002;64(3):479-498. doi:10.1111/1467-9868.00346 

228. Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic Brain Injury–Related Emergency 

Department Visits, Hospitalizations, and Deaths — United States, 2007 and 2013. MMWR 
Surveill Summ. 2017;66(9):1-16. doi:10.15585/mmwr.ss6609a1 

229. Centers for Disease Control and Prevention. (2018). Report to Congress: The Management of 

Traumatic Brain Injury in Children, National Center for Injury Prevention and Control; Division 
of Unintentional Injury Prevention. Atlanta, GA.   

230. Muftuler LT, Meier TB, Keith M, Budde MD, Huber DL, McCrea MA. Serial Diffusion 

Kurtosis Magnetic Resonance Imaging Study during Acute, Subacute, and Recovery Periods 
after Sport-Related Concussion. J Neurotrauma. 2020;37(19):2081-2092. 
doi:10.1089/neu.2020.6993 

231. Slobounov SM, Walter A, Breiter HC, et al. The effect of repetitive subconcussive collisions on 

brain integrity in collegiate football players over a single football season: A multi-modal 
neuroimaging study. NeuroImage Clin. 2017;14:708-718. doi:10.1016/j.nicl.2017.03.006 

232. Ooi LQR, Orban C, Nichols T, et al. MRI Economics: Balancing Sample Size and Scan Duration in 

Brain Wide Association Studies. Neuroscience; 2024. doi:10.1101/2024.02.16.580448 

233. Scarpazza C, Ha M, Baecker L, et al. Translating research findings into clinical practice: a 

systematic and critical review of neuroimaging-based clinical tools for brain disorders. Transl 
Psychiatry. 2020;10(1):107. doi:10.1038/s41398-020-0798-6 

 150 

234. Wintermark M, Colen R, Whitlow CT, Zaharchuk G. The vast potential and bright future of 
neuroimaging. Br J Radiol. Published online June 6, 2018:20170505. doi:10.1259/bjr.20170505 

235. Echemendia RJ, Meeuwisse W, McCrory P, et al. The Sport Concussion Assessment Tool 5th 
Edition (SCAT5). Br J Sports Med. Published online April 26, 2017:bjsports-2017-097506. 
doi:10.1136/bjsports-2017-097506 

236. Dienes Z. Bayesian Versus Orthodox Statistics: Which Side Are You On? Perspect Psychol Sci. 

2011;6(3):274-290. doi:10.1177/1745691611406920 

237. Afzali M, Pieciak T, Newman S, et al. The sensitivity of diffusion MRI to microstructural 

properties and experimental factors. J Neurosci Methods. 2021;347:108951. 
doi:10.1016/j.jneumeth.2020.108951 

238. Setsompop K, Kimmlingen R, Eberlein E, et al. Pushing the limits of in vivo diffusion MRI for 

the Human Connectome Project. NeuroImage. 2013;80:220-233. 
doi:10.1016/j.neuroimage.2013.05.078 

239. Bennett CM, Wolford GL, Miller MB. The principled control of false positives in 
neuroimaging. Soc Cogn Affect Neurosci. 2009;4(4):417-422. doi:10.1093/scan/nsp053 

240. Mohammed FN, Master CL, Arbogast KB, et al. Disparities in Adherence to Concussion 

Clinical Care Recommendations in a Pediatric Population. J Head Trauma Rehabil. 
2023;38(2):147-155. doi:10.1097/HTR.0000000000000823 

241. Seabury SA, Gaudette É, Goldman DP, et al. Assessment of Follow-up Care After Emergency 
Department Presentation for Mild Traumatic Brain Injury and Concussion: Results From the 
TRACK-TBI Study. JAMA Netw Open. 2018;1(1):e180210. 
doi:10.1001/jamanetworkopen.2018.0210 

242. Merritt VC, Padgett CR, Jak AJ. A systematic review of sex differences in concussion outcome: 

What do we know? Clin Neuropsychol. 2019;33(6):1016-1043. 
doi:10.1080/13854046.2018.1508616 

243. Wiederschain GYa. Data mining techniques for the life sciences: (Olivero Carugo and Frank 
Eisenhaber (eds.), in Series “Springer Protocols. Methods in Molecular Biology”, Vol. 609, 
Humana Press, 2010, 407 p., $110). Biochem Mosc. 2011;76(4):494-494. 
doi:10.1134/S0006297911040158 

244. Vakil E, Blachstein H. Rey auditory-verbal learning test: Structure analysis. J Clin Psychol. 

1993;49(6):883-890. doi:10.1002/1097-4679(199311)49:6<883::AID-
JCLP2270490616>3.0.CO;2-6 

245. Huettel SA, Song AW, McCarthy G. Functional Magnetic Resonance Imaging. Sinauer Associates, 

Publishers; 2009. 

246. Kochanek PM, Hendrich KS, Dixon CE, Schiding JK, Williams DS, Ho C. Cerebral Blood 

Flow at One Year after Controlled Cortical Impact in Rats: Assessment by Magnetic Resonance 
Imaging. J Neurotrauma. 2002;19(9):1029-1037. doi:10.1089/089771502760341947 

 151 

247. Hendrich KS, Kochanek PM, Williams DS, Schiding JK, Marion DW, Ho C. Early perfusion 
after controlled cortical impact in rats: Quantification by arterial spin-labeled MRI and the 
influence of spin-lattice relaxation time heterogeneity. Magn Reson Med. 1999;42(4):673-681. 
doi:10.1002/(SICI)1522-2594(199910)42:4<673::AID-MRM8>3.0.CO;2-B 

248. DeWitt DS, Prough DS. Traumatic Cerebral Vascular Injury: The Effects of Concussive Brain 

Injury on the Cerebral Vasculature. J Neurotrauma. 2003;20(9):795-825. 
doi:10.1089/089771503322385755 

249. Meier TB, Bellgowan PSF, Singh R, Kuplicki R, Polanski DW, Mayer AR. Recovery of Cerebral 

Blood Flow Following Sports-Related Concussion. JAMA Neurol. 2015;72(5):530. 
doi:10.1001/jamaneurol.2014.4778 

250. Kong L, Qiu S, Chen Y, et al. Assessment of vibration modulated regional cerebral blood flow 

with MRI. NeuroImage. 2023;269:119934. doi:10.1016/j.neuroimage.2023.119934 

251. Wang Y, Nencka AS, Meier TB, et al. Cerebral blood flow in acute concussion: preliminary 
ASL findings from the NCAA-DoD CARE consortium. Brain Imaging Behav. 2019;13(5):1375-
1385. doi:10.1007/s11682-018-9946-5 

252. Brett BL, Cohen AD, McCrea MA, Wang Y. Longitudinal alterations in cerebral perfusion 

following a season of adolescent contact sport participation compared to non-contact athletes. 
NeuroImage Clin. 2023;40:103538. doi:10.1016/j.nicl.2023.103538 

253. Wang Y, Bartels HM, Nelson LD. A Systematic Review of ASL Perfusion MRI in Mild TBI. 

Neuropsychol Rev. 2023;33(1):160-191. doi:10.1007/s11065-020-09451-7 

254. Ramsey S, Willke R, Briggs A, et al. Good Research Practices for Cost-Effectiveness Analysis 

Alongside Clinical Trials: The ISPOR RCT-CEA Task Force Report. Value Health. 
2005;8(5):521-533. doi:10.1111/j.1524-4733.2005.00045.x 

255. Chen WL, Wagner J, Heugel N, et al. Functional Near-Infrared Spectroscopy and Its Clinical 
Application in the Field of Neuroscience: Advances and Future Directions. Front Neurosci. 
2020;14:724. doi:10.3389/fnins.2020.00724 

256. Tsow F, Kumar A, Hosseini SH, Bowden A. A low-cost, wearable, do-it-yourself functional 

near-infrared spectroscopy (DIY-fNIRS) headband. HardwareX. 2021;10:e00204. 
doi:10.1016/j.ohx.2021.e00204 

257. Helmich I, Saluja RS, Lausberg H, et al. Persistent Postconcussive Symptoms Are 

Accompanied by Decreased Functional Brain Oxygenation. J Neuropsychiatry Clin Neurosci. 
2015;27(4):287-298. doi:10.1176/appi.neuropsych.14100276 

258. Grijalva C, Mullins VA, Michael BR, et al. Neuroimaging, wearable sensors, and blood-based 

biomarkers reveal hyperacute changes in the brain after sub-concussive impacts. Brain 
Multiphysics. 2023;5:100086. doi:10.1016/j.brain.2023.100086 

 152 

 
 
 
APPENDIX 

Joint Caption for Appendix Tables A.1-14- All tables are t-test results for comparisons of the 
automatically segmented tracts from TractSeg between mTBI and control subjects.  

Joint Caption for Appendix Figures A.1-94 – All figures are boxplots representing significant 
comparisons from appendix tables 1-14. Boxplots are colored green if the false discovery rate 
corrected q-value was q<0.05, and otherwise appear black. Titles are the matching abbreviated 
TractSeg labels from Tables 1-14, with the abbreviated metric being compared. Y-labels indicate the 
signal representation or modelling technique used to derive the metric and the full name of the 
metric as in the abbreviated title. The results from 2-week timepoint (P1), 6-month timepoint (P2) 
are arranged with P1 on the left and P2 on the right.  

Joint Caption for Appendix Tables A.15-25- All tables are t-test results for comparisons of the 
automatically segmented tracts from TractSeg between timepoints within mTBI and control subject 
groups.  

 153 

 
 
 
 
Table A.1 Results of Post Hoc Tract Specific Comparisons of Diffusion Tensor Imaging 
Fractional Anisotropy 

mTBI 

Controls 

Percent Change    

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  Contr
ols 

t-score 
P1 

q-value 
P11 

t-
score 
P2 

q-
value 
P21 

AF_left 

0.256 ± 0.013 

0.256 ± 0.011 

0.259 ± 0.019 

0.260 ± 0.016 

-0.12 % 

AF_right 

0.267 ± 0.011 

0.266 ± 0.013 

0.263 ± 0.016 

0.261 ± 0.026 

-0.45 % 

ATR_left 

0.271 ± 0.011 

0.271 ± 0.012 

0.278 ± 0.020 

0.280 ± 0.019 

-0.03 % 

0.41 
% 

-0.64 
% 

0.74 
% 

-0.716 

0.875 

-1.238 

0.472 

1.248 

0.875 

1.053 

0.472 

-1.926 

0.494 

-2.521 

0.271 

ATR_right 

0.263 ± 0.010 

0.264 ± 0.012 

0.266 ± 0.016 

0.269 ± 0.018 

0.3 % 

1.3 % 

-0.844 

0.875 

-1.418 

0.472 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

0.270 ± 0.017 

0.267 ± 0.022 

0.287 ± 0.024 

0.286 ± 0.030 

-1.11 % 

-0.4 % 

-3.227 

0.119 

-2.827 

0.237 

0.343 ± 0.023 

0.336 ± 0.032 

0.351 ± 0.023 

0.343 ± 0.056 

-1.88 % 

0.296 ± 0.014 

0.295 ± 0.013 

0.298 ± 0.023 

0.298 ± 0.025 

-0.27 % 

0.291 ± 0.017 

0.290 ± 0.021 

0.295 ± 0.035 

0.295 ± 0.034 

-0.47 % 

0.307 ± 0.018 

0.305 ± 0.019 

0.305 ± 0.032 

0.307 ± 0.031 

-0.63 % 

-2.17 
% 

-0.03 
% 

0.16 
% 

0.83 
% 

-1.195 

0.875 

-0.613 

0.472 

-0.411 

0.917 

-0.560 

0.472 

-0.615 

0.916 

-0.831 

0.472 

0.386 

0.917 

-0.341 

0.478 

0.318 ± 0.017 

0.314 ± 0.020 

0.313 ± 0.033 

0.315 ± 0.035 

-1.12 %  0.6 % 

0.742 

0.875 

-0.128 

0.525 

0.301 ± 0.016 

0.300 ± 0.015 

0.295 ± 0.023 

0.293 ± 0.026 

-0.25 % 

-0.57 
% 

1.234 

0.875 

1.435 

0.472 

0.308 ± 0.015 

0.306 ± 0.016 

0.304 ± 0.014 

0.297 ± 0.037 

-0.57 % 

-2.3 % 

0.797 

0.298 ± 0.013 

0.297 ± 0.013 

0.296 ± 0.022 

0.294 ± 0.025 

-0.27 % 

0.429 

0.875 

0.917 

1.401 

0.522 

0.472 

0.472 

CG_left 

0.278 ± 0.017 

0.281 ± 0.021 

0.279 ± 0.023 

0.277 ± 0.027  1.02 % 

-0.180 

0.922 

0.704 

0.472 

CG_right 

CST_left 

0.282 ± 0.019 

0.283 ± 0.016 

0.280 ± 0.025 

0.277 ± 0.025  0.27 % 

-1 % 

0.457 

0.917 

1.241 

0.472 

0.344 ± 0.014 

0.345 ± 0.016 

0.347 ± 0.025 

0.351 ± 0.024  0.46 % 

-0.43 
% 

-0.92 
% 

1.16 
% 

-0.01 
% 

0.65 
% 

1.01 
% 

-1.51 
% 

1.95 
% 

1.02 
% 

2.42 
% 

-1.94 
% 

-0.63 
% 

-3.15 
% 

-2.31 
% 

0.02 
% 

-0.557 

0.916 

-1.022 

0.472 

0.385 

0.917 

0.262 

0.485 

-0.263 

0.921 

-0.345 

0.478 

0.250 

0.921 

-0.472 

0.472 

-0.510 

0.917 

-0.683 

0.472 

-0.051 

0.924 

-0.784 

0.472 

1.956 

0.494 

1.079 

0.472 

3.021 

0.119 

1.155 

0.472 

-0.652 

0.9 

0.417 

0.473 

-0.005 

0.94 

0.094 

0.532 

-1.777 

0.603 

0.213 

0.493 

-1.376 

0.875 

0.005 

0.541 

1.934 

0.494 

1.242 

0.472 

CST_right 

0.352 ± 0.013 

0.351 ± 0.017 

0.350 ± 0.025 

0.350 ± 0.031 

-0.08 % 

FPT_left 

0.333 ± 0.014 

0.335 ± 0.015 

0.334 ± 0.027 

0.336 ± 0.027  0.52 % 

FPT_right 

0.333 ± 0.013 

0.333 ± 0.014 

0.331 ± 0.023 

0.335 ± 0.025 

0 % 

FX_left 

0.288 ± 0.046 

0.281 ± 0.045 

0.295 ± 0.047 

0.291 ± 0.055 

-2.34 % 

FX_right 

0.294 ± 0.045 

0.289 ± 0.049 

0.294 ± 0.048 

0.300 ± 0.052 

-1.54 % 

ICP_left 

0.275 ± 0.025 

0.273 ± 0.030 

0.260 ± 0.029 

0.263 ± 0.044 

-0.62 % 

ICP_right 

0.258 ± 0.016 

0.255 ± 0.017 

0.242 ± 0.025 

0.248 ± 0.031 

-1.4 % 

IFO_left 

0.283 ± 0.011 

0.281 ± 0.011 

0.285 ± 0.013 

0.279 ± 0.029 

-0.51 % 

IFO_right 

0.281 ± 0.010 

0.279 ± 0.012 

0.281 ± 0.011 

0.279 ± 0.027 

-0.47 % 

ILF_left 

0.277 ± 0.011 

0.275 ± 0.017 

0.283 ± 0.012 

0.274 ± 0.032 

-0.7 % 

ILF_right 

0.273 ± 0.011 

0.272 ± 0.019 

0.278 ± 0.013 

0.272 ± 0.033 

-0.66 % 

MCP 

0.301 ± 0.015 

0.299 ± 0.016 

0.291 ± 0.029 

0.291 ± 0.037 

-0.87 % 

 154 

 
 
 
 
 
Table A.1 (cont’d) 

mTBI 

Controls 

Percent Change 

TractSeg Abbv 

P1 

P2 

P1 

P2 

mTBI  Contr
ols 

t-score 
P1 

t-score 
P2 

q-
value 
P1 

q-
value 
P2 

MLF_left 

0.265 ± 0.014 

0.266 ± 0.014 

0.262 ± 0.018 

0.264 ± 0.016  0.39 % 

MLF_right 

0.272 ± 0.013 

0.271 ± 0.014 

0.266 ± 0.020 

0.264 ± 0.023 

-0.35 % 

OR_left 

0.293 ± 0.014 

0.292 ± 0.014 

0.294 ± 0.009 

0.289 ± 0.035 

-0.43 % 

OR_right 

0.300 ± 0.013 

0.300 ± 0.015 

0.300 ± 0.015 

0.297 ± 0.026 

-0.26 % 

POPT_left 

0.313 ± 0.013 

0.313 ± 0.015 

0.309 ± 0.022 

0.311 ± 0.022  0.14 % 

POPT_right 

0.327 ± 0.012 

0.326 ± 0.016 

0.323 ± 0.023 

0.325 ± 0.024 

-0.16 % 

SCP_left 

0.313 ± 0.019 

0.313 ± 0.023 

0.308 ± 0.026 

0.315 ± 0.034 

-0.01 % 

SCP_right 

0.313 ± 0.013 

0.311 ± 0.015 

0.302 ± 0.029 

0.310 ± 0.028 

-0.66 % 

SLF_III_left 

0.264 ± 0.015 

0.262 ± 0.016 

0.263 ± 0.028 

0.266 ± 0.022 

-0.76 % 

SLF_III_right 

0.273 ± 0.012 

0.270 ± 0.020 

0.263 ± 0.020 

0.259 ± 0.038 

-1.15 % 

SLF_II_left 

0.250 ± 0.015 

0.253 ± 0.016 

0.249 ± 0.022 

0.255 ± 0.027  1.16 % 

SLF_II_right 

0.252 ± 0.014 

0.253 ± 0.016 

0.247 ± 0.021 

0.249 ± 0.018 

0.5 % 

SLF_I_left 

0.243 ± 0.014 

0.245 ± 0.018 

0.238 ± 0.028 

0.239 ± 0.023 

1 % 

SLF_I_right 

0.257 ± 0.014 

0.259 ± 0.019 

0.252 ± 0.027 

0.256 ± 0.022  0.97 % 

STR_left 

0.335 ± 0.017 

0.336 ± 0.019 

0.338 ± 0.024 

0.338 ± 0.035  0.47 % 

STR_right 

0.339 ± 0.016 

0.339 ± 0.020 

0.339 ± 0.027 

0.344 ± 0.023 

-0.13 % 

ST_FO_left 

0.285 ± 0.017 

0.281 ± 0.018 

0.295 ± 0.016 

0.295 ± 0.033 

-1.41 % 

ST_FO_right 

0.290 ± 0.017 

0.288 ± 0.020 

0.294 ± 0.016 

0.293 ± 0.037 

-0.75 % 

ST_OCC_left 

0.294 ± 0.014 

0.293 ± 0.013 

0.294 ± 0.013 

0.289 ± 0.036 

-0.33 % 

0.57 
% 

-0.71 
% 

-1.73 
% 

-0.99 
% 

0.84 
% 

0.52 
% 

2.26 
% 

2.62 
% 

1.03 
% 

-1.43 
% 

2.16 
% 

0.96 
% 

0.68 
% 

1.65 
% 

0.18 
% 

1.42 
% 

0.09 
% 

-0.36 
% 

-1.62 
% 

0.585 

0.916 

0.512 

0.472 

1.480 

0.875 

1.569 

0.472 

-0.221 

0.921 

0.525 

0.472 

0.225 

0.921 

0.623 

0.472 

0.988 

0.875 

0.436 

0.472 

0.767 

0.875 

0.228 

0.493 

0.778 

0.875 

-0.323 

0.478 

2.078 

0.494 

0.077 

0.532 

0.183 

0.922 

-0.774 

0.472 

2.566 

0.281 

1.587 

0.472 

0.107 

0.922 

-0.371 

0.478 

1.145 

0.875 

0.873 

0.472 

0.990 

0.875 

1.093 

0.472 

0.913 

0.875 

0.556 

0.472 

-0.568 

0.916 

-0.303 

0.478 

0.098 

0.922 

-0.825 

0.472 

-2.075 

0.494 

-2.306 

0.313 

-0.852 

0.875 

-0.743 

0.472 

0.050 

0.924 

0.702 

0.472 

 155 

 
 
 
 
 
 
Table A.1 (cont’d) 

TractSeg Abbv 

P1 

P2 

P1 

P2 

mTBI  Controls 

mTBI 

Controls  

Percent Change 

t-
score 
P1 

q-
value 
P1 

t-
score 
P2 

q-
value 
P2 

ST_PAR_left 

0.280 ± 0.012 

0.282 ± 0.016 

0.278 ± 0.019 

0.279 ± 0.018 

0.46 % 

0.68 % 

0.668 

0.9 

0.462 

0.472 

ST_PAR_right 

0.296 ± 0.012 

0.295 ± 0.015 

0.292 ± 0.019 

0.293 ± 0.020 

-0.22 % 

0.21 % 

0.938 

0.875 

0.540 

0.472 

ST_POSTC_left 

0.291 ± 0.014 

0.292 ± 0.018 

0.292 ± 0.019 

0.296 ± 0.015 

0.42 % 

1.2 % 

-0.305  0.921 

-0.744  0.472 

ST_POSTC_right 

0.308 ± 0.014 

0.306 ± 0.020 

0.307 ± 0.022 

0.308 ± 0.023 

-0.5 % 

0.62 % 

0.245 

0.921 

-0.402  0.473 

ST_PREC_left 

0.295 ± 0.012 

0.296 ± 0.015 

0.300 ± 0.019 

0.301 ± 0.017 

0.59 % 

0.61 % 

-1.189  0.875 

-1.106  0.472 

ST_PREC_right 

0.305 ± 0.012 

0.305 ± 0.017 

0.305 ± 0.019 

0.307 ± 0.019 

-0.07 % 

0.81 % 

0.132 

0.922 

-0.447  0.472 

ST_PREF_left 

0.285 ± 0.012 

0.286 ± 0.011 

0.289 ± 0.020 

0.288 ± 0.020 

0.38 % 

-0.36 % 

-1.111  0.875 

-0.597  0.472 

ST_PREF_right 

0.283 ± 0.010 

0.283 ± 0.011 

0.284 ± 0.014 

0.286 ± 0.019 

-0.05 % 

0.68 % 

-0.292  0.921 

-0.814  0.472 

ST_PREM_left 

0.272 ± 0.011 

0.274 ± 0.011 

0.275 ± 0.022 

0.277 ± 0.018 

0.65 % 

0.5 % 

-0.736  0.875 

-0.697  0.472 

ST_PREM_right 

0.285 ± 0.011 

0.285 ± 0.013 

0.285 ± 0.021 

0.285 ± 0.024 

0.24 % 

0.16 % 

-0.044  0.924 

0.012 

0.541 

T_OCC_left 

0.289 ± 0.013 

0.288 ± 0.014 

0.291 ± 0.009 

0.285 ± 0.034 

-0.4 % 

-1.79 % 

-0.405  0.917 

0.466 

0.472 

T_OCC_right 

0.295 ± 0.013 

0.294 ± 0.014 

0.294 ± 0.014 

0.291 ± 0.025 

-0.18 % 

-0.98 % 

0.155 

0.922 

0.628 

0.472 

T_PAR_left 

0.280 ± 0.012 

0.282 ± 0.016 

0.277 ± 0.019 

0.279 ± 0.020 

0.45 % 

0.71 % 

0.754 

0.875 

0.486 

0.472 

T_PAR_right 

0.296 ± 0.013 

0.296 ± 0.016 

0.292 ± 0.020 

0.294 ± 0.020 

0.13 % 

0.5 % 

0.887 

0.875 

0.568 

0.472 

T_POSTC_left 

0.296 ± 0.014 

0.297 ± 0.018 

0.293 ± 0.025 

0.299 ± 0.022 

0.22 % 

2.07 % 

0.795 

0.875 

-0.294  0.478 

T_POSTC_right 

0.311 ± 0.016 

0.310 ± 0.023 

0.305 ± 0.033 

0.307 ± 0.029 

-0.27 % 

0.64 % 

0.993 

0.875 

0.468 

0.472 

T_PREC_left 

0.300 ± 0.013 

0.302 ± 0.017 

0.304 ± 0.023 

0.307 ± 0.022 

0.66 % 

1.05 % 

-0.969  0.875 

-1.101  0.472 

T_PREC_right 

0.304 ± 0.013 

0.305 ± 0.018 

0.302 ± 0.024 

0.305 ± 0.020 

0.47 % 

0.96 % 

0.263 

0.921 

-0.057  0.533 

T_PREF_left 

0.287 ± 0.012 

0.289 ± 0.013 

0.292 ± 0.022 

0.292 ± 0.022 

0.55 % 

0.11 % 

-1.199  0.875 

-0.885  0.472 

T_PREF_right 

0.277 ± 0.011 

0.278 ± 0.013 

0.278 ± 0.017 

0.282 ± 0.016 

0.4 % 

1.23 % 

-0.360  0.921 

-0.945  0.472 

T_PREM_left 

0.283 ± 0.013 

0.284 ± 0.015 

0.287 ± 0.023 

0.292 ± 0.021 

0.48 % 

1.63 % 

-0.980  0.875 

-1.675  0.472 

T_PREM_right 

0.286 ± 0.014 

0.289 ± 0.016 

0.288 ± 0.027 

0.291 ± 0.023 

0.92 % 

1.01 % 

-0.442  0.917 

-0.505  0.472 

UF_left 

0.277 ± 0.014 

0.273 ± 0.017 

0.281 ± 0.019 

0.277 ± 0.029 

-1.6 % 

-1.17 % 

-0.744  0.875 

-0.784  0.472 

0.275 ± 0.013 

0.271 ± 0.018 

UF_right 
Tables A.1- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

-0.470  0.917 

0.276 ± 0.033 

0.277 ± 0.014 

-0.37 % 

-1.45 % 

-0.760  0.472 

 156 

 
 
 
 
 
 
 
 
 
Table A.2 Results of Post Hoc Tract Specific Comparisons of Diffusion Tensor Imaging Mean 
Diffusivity  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value P21 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

-0.7 % 

-0.91 % 

3.362 

0.007* 

3.522 

0.001* 

-0.4 % 

-1.01 % 

1.560 

0.138 

2.567 

0.009* 

-0.61 % 

-1.91 % 

2.397 

0.043* 

3.114 

0.004* 

-0.95 % 

-1.86 % 

1.068 

0.241 

1.837 

0.028* 

-0.33 % 

-1.95 % 

5.569 

<0.001 

5.289 

<0.001 

-0.48 % 

-2.96 % 

0.668 

0.336 

2.335 

0.012* 

-0.02 % 

-1.45 % 

1.909 

0.09 

3.036 

0.004* 

0.29 % 

-0.97 % 

1.789 

0.107 

2.066 

0.02* 

0.58 % 

0.61 % 

1.115 

0.241 

0.700 

0.109 

1.04 % 

0.61 % 

0.744 

0.316 

0.668 

0.109 

0.41 % 

0.59 % 

0.627 

0.338 

0.428 

0.133 

0.02 % 

0.81 % 

0.353 

0.422 

-0.096 

0.164 

0.19 % 

-0.25 % 

1.079 

0.241 

1.293 

0.06 

-0.17 % 

0.49 % 

1.622 

0.126 

0.721 

0.109 

-0.32 % 

-0.52 % 

1.299 

0.197 

1.414 

0.051 

-0.25 % 

-0.58 % 

3.140 

0.011* 

2.377 

0.011* 

-0.27 % 

0.99 % 

2.492 

0.043* 

0.831 

0.103 

-0.14 % 

-1.61 % 

2.278 

0.05 

2.986 

0.004* 

-0.13 % 

-1.85 % 

1.721 

0.11 

3.122 

0.004* 

-1.21 % 

-1.03 % 

0.181 

0.48 

0.137 

0.162 

-0.9 % 

-3.5 % 

0.129 

0.48 

0.684 

0.109 

-0.81 % 

-1.14 % 

0.298 

0.436 

0.596 

0.116 

-0.64 % 

-3.28 % 

-0.117 

0.48 

1.600 

0.039* 

-0.12 % 

-0.37 % 

2.476 

0.043* 

2.524 

0.009* 

-0.32 % 

-0.74 % 

1.317 

0.197 

2.027 

0.02* 

-0.8 % 

-0.89 % 

2.136 

0.062 

2.457 

0.01* 

-0.64 % 

-1.04 % 

1.096 

0.241 

1.635 

0.037* 

7.999e-04 ± 
2.322e-05 

7.868e-04 ± 
2.222e-05 

8.409e-04 ± 
3.346e-05 

8.461e-04 ± 
3.803e-05 

8.916e-04 ± 
2.541e-05 

8.586e-04 ± 
4.152e-05 

8.375e-04 ± 
2.624e-05 

8.569e-04 ± 
3.301e-05 

8.417e-04 ± 
2.701e-05 

8.335e-04 ± 
2.607e-05 

8.171e-04 ± 
3.205e-05 

8.307e-04 ± 
5.528e-05 

8.297e-04 ± 
2.690e-05 

7.949e-04 ± 
2.359e-05 

7.884e-04 ± 
2.595e-05 

8.246e-04 ± 
2.229e-05 

7.904e-04 ± 
2.320e-05 

8.283e-04 ± 
2.218e-05 

8.233e-04 ± 
2.398e-05 

1.609e-03 ± 
2.432e-04 

1.594e-03 ± 
2.327e-04 

7.459e-04 ± 
5.752e-05 

7.377e-04 ± 
4.462e-05 

8.088e-04 ± 
2.674e-05 

8.034e-04 ± 
3.017e-05 

8.232e-04 ± 
3.840e-05 

8.137e-04 ± 
3.955e-05 

7.943e-04 ± 
2.429e-05 

7.756e-04 ± 
3.105e-05 

7.837e-04 ± 
1.881e-05 

7.764e-04 ± 
2.585e-05 

8.358e-04 ± 
3.916e-05 

8.161e-04 ± 
4.401e-05 

8.380e-04 ± 
3.574e-05 

8.330e-04 ± 
5.677e-05 

8.887e-04 ± 
4.040e-05 

8.460e-04 ± 
3.763e-05 

8.545e-04 ± 
4.578e-05 

8.494e-04 ± 
6.718e-05 

8.373e-04 ± 
2.961e-05 

8.216e-04 ± 
3.730e-05 

8.594e-04 ± 
4.634e-05 

8.383e-04 ± 
4.583e-05 

8.466e-04 ± 
3.477e-05 

8.317e-04 ± 
4.394e-05 

8.422e-04 ± 
3.891e-05 

8.266e-04 ± 
4.895e-05 

8.205e-04 ± 
2.704e-05 

8.113e-04 ± 
3.214e-05 

8.309e-04 ± 
3.870e-05 

8.254e-04 ± 
3.388e-05 

8.312e-04 ± 
2.563e-05 

8.208e-04 ± 
3.323e-05 

7.935e-04 ± 
2.323e-05 

7.818e-04 ± 
4.112e-05 

7.859e-04 ± 
2.197e-05 

7.774e-04 ± 
4.034e-05 

8.226e-04 ± 
3.099e-05 

7.999e-04 ± 
4.135e-05 

7.882e-04 ± 
2.337e-05 

7.718e-04 ± 
3.429e-05 

8.271e-04 ± 
3.052e-05 

8.068e-04 ± 
5.824e-05 

8.223e-04 ± 
2.555e-05 

8.074e-04 ± 
5.361e-05 

1.590e-03 ± 
2.420e-04 

1.597e-03 ± 
2.552e-04 

1.579e-03 ± 
2.466e-04 

1.585e-03 ± 
2.478e-04 

7.398e-04 ± 
3.820e-05 

7.409e-04 ± 
5.451e-05 

7.329e-04 ± 
2.946e-05 

7.393e-04 ± 
5.818e-05 

8.078e-04 ± 
2.595e-05 

7.904e-04 ± 
2.080e-05 

8.009e-04 ± 
2.404e-05 

7.924e-04 ± 
2.364e-05 

8.165e-04 ± 
3.108e-05 

8.006e-04 ± 
2.704e-05 

8.085e-04 ± 
2.961e-05 

8.014e-04 ± 
3.538e-05 

7.686e-04 ± 
3.164e-05 

7.686e-04 ± 
2.719e-05 

8.005e-04 ± 
4.388e-05 

8.175e-04 ± 
5.208e-05 

8.296e-04 ± 
3.660e-05 

8.243e-04 ± 
4.685e-05 

8.096e-04 ± 
4.104e-05 

8.302e-04 ± 
6.241e-05 

8.367e-04 ± 
8.721e-05 

8.317e-04 ± 
9.654e-05 

8.161e-04 ± 
6.004e-05 

8.321e-04 ± 
6.310e-05 

8.188e-04 ± 
5.572e-05 

7.856e-04 ± 
7.225e-05 

7.734e-04 ± 
5.455e-05 

7.953e-04 ± 
6.590e-05 

7.795e-04 ± 
6.850e-05 

7.937e-04 ± 
6.377e-05 

7.924e-04 ± 
5.552e-05 

1.580e-03 ± 
2.864e-04 

1.529e-03 ± 
3.021e-04 

7.325e-04 ± 
5.920e-05 

7.150e-04 ± 
6.579e-05 

7.875e-04 ± 
3.704e-05 

7.865e-04 ± 
2.874e-05 

7.935e-04 ± 
4.082e-05 

7.931e-04 ± 
4.544e-05 

 157 

 
 
 
 
 
 
 
Table A.2 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  Controls 

mTBI 

Controls 

Percent Change 

q-value 
P11 

t-score 
P2 

q-value P21 

t-
score 
P1 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

7.251e-04 ± 
4.452e-05 

8.097e-04 ± 
2.705e-05 

8.024e-04 ± 
2.876e-05 

8.340e-04 ± 
3.934e-05 

8.257e-04 ± 
4.445e-05 

8.315e-04 ± 
2.473e-05 

8.162e-04 ± 
2.751e-05 

7.868e-04 ± 
2.581e-05 

7.779e-04 ± 
2.147e-05 

7.962e-04 ± 
2.382e-05 

7.828e-04 ± 
2.264e-05 

8.100e-04 ± 
2.939e-05 

8.013e-04 ± 
2.816e-05 

8.457e-04 ± 
3.854e-05 

8.284e-04 ± 
3.527e-05 

7.828e-04 ± 
2.238e-05 

7.658e-04 ± 
2.186e-05 

8.061e-04 ± 
3.489e-05 

8.106e-04 ± 
3.320e-05 

8.169e-04 ± 
3.222e-05 

8.034e-04 ± 
3.557e-05 

8.232e-04 ± 
2.826e-05 

8.055e-04 ± 
2.877e-05 

8.273e-04 ± 
2.845e-05 

7.981e-04 ± 
2.772e-05 

8.188e-04 ± 
2.658e-05 

7.950e-04 ± 
2.800e-05 

7.239e-04 ± 
3.381e-05 

7.178e-04 ± 
5.040e-05 

8.088e-04 ± 
2.591e-05 

8.013e-04 ± 
3.068e-05 

8.033e-04 ± 
2.571e-05 

7.993e-04 ± 
3.606e-05 

8.315e-04 ± 
3.878e-05 

8.193e-04 ± 
3.161e-05 

8.239e-04 ± 
3.876e-05 

8.261e-04 ± 
3.738e-05 

8.350e-04 ± 
2.884e-05 

8.286e-04 ± 
4.189e-05 

8.184e-04 ± 
2.682e-05 

8.094e-04 ± 
3.952e-05 

7.823e-04 ± 
3.232e-05 

7.625e-04 ± 
4.451e-05 

7.733e-04 ± 
2.205e-05 

7.682e-04 ± 
5.620e-05 

7.944e-04 ± 
2.694e-05 

7.806e-04 ± 
5.033e-05 

7.829e-04 ± 
2.113e-05 

7.793e-04 ± 
3.347e-05 

8.057e-04 ± 
3.014e-05 

7.989e-04 ± 
4.336e-05 

7.979e-04 ± 
2.601e-05 

7.935e-04 ± 
4.066e-05 

8.478e-04 ± 
4.422e-05 

8.576e-04 ± 
8.340e-05 

8.267e-04 ± 
3.849e-05 

8.296e-04 ± 
6.358e-05 

7.788e-04 ± 
3.467e-05 

7.458e-04 ± 
4.054e-05 

7.614e-04 ± 
2.653e-05 

7.499e-04 ± 
4.676e-05 

8.052e-04 ± 
3.935e-05 

7.719e-04 ± 
3.447e-05 

8.071e-04 ± 
3.231e-05 

7.906e-04 ± 
4.002e-05 

8.147e-04 ± 
3.153e-05 

7.958e-04 ± 
1.883e-05 

8.012e-04 ± 
3.058e-05 

7.937e-04 ± 
2.386e-05 

8.248e-04 ± 
2.954e-05 

8.176e-04 ± 
3.713e-05 

8.060e-04 ± 
2.623e-05 

7.978e-04 ± 
3.132e-05 

8.256e-04 ± 
3.665e-05 

8.107e-04 ± 
4.887e-05 

7.965e-04 ± 
2.909e-05 

7.923e-04 ± 
4.763e-05 

8.139e-04 ± 
3.580e-05 

7.880e-04 ± 
4.243e-05 

7.895e-04 ± 
2.563e-05 

7.793e-04 ± 
3.716e-05 

7.300e-04 ± 
6.087e-05 

7.968e-04 ± 
4.225e-05 

7.949e-04 ± 
4.326e-05 

8.179e-04 ± 
6.352e-05 

8.206e-04 ± 
5.220e-05 

8.256e-04 ± 
7.224e-05 

8.114e-04 ± 
6.759e-05 

7.543e-04 ± 
5.961e-05 

7.523e-04 ± 
3.878e-05 

7.732e-04 ± 
5.376e-05 

7.740e-04 ± 
4.087e-05 

7.891e-04 ± 
5.031e-05 

7.839e-04 ± 
4.224e-05 

8.492e-04 ± 
9.782e-05 

8.219e-04 ± 
8.344e-05 

7.504e-04 ± 
9.970e-05 

7.410e-04 ± 
6.010e-05 

7.685e-04 ± 
5.548e-05 

7.746e-04 ± 
2.859e-05 

7.951e-04 ± 
4.364e-05 

7.912e-04 ± 
3.384e-05 

8.152e-04 ± 
6.397e-05 

7.990e-04 ± 
5.421e-05 

8.032e-04 ± 
7.729e-05 

7.963e-04 ± 
7.827e-05 

7.863e-04 ± 
6.738e-05 

7.789e-04 ± 
5.685e-05 

 158 

-0.17 % 

1.7 % 

0.551 

0.36 

-0.528 

0.121 

-0.1 % 

-0.56 % 

1.041 

0.246 

1.427 

0.051 

0.12 % 

-0.55 % 

0.346 

0.422 

0.984 

0.086 

-0.3 % 

-0.17 % 

1.342 

0.197 

1.075 

0.078 

-0.22 % 

-0.66 % 

-0.034 

0.491 

0.273 

0.149 

0.41 % 

-0.36 % 

0.351 

0.422 

0.801 

0.105 

0.28 % 

0.24 % 

0.770 

0.312 

0.644 

0.111 

-0.58 % 

-1.08 % 

2.762 

0.029* 

2.515 

0.009* 

-0.6 % 

-2.06 % 

1.070 

0.241 

2.821 

0.005* 

-0.23 % 

-0.95 % 

1.736 

0.11 

2.192 

0.015* 

0.01 % 

-0.68 % 

0.477 

0.386 

1.187 

0.07 

-0.54 % 

-1.22 % 

1.176 

0.229 

1.661 

0.037* 

-0.43 % 

-1.22 % 

0.869 

0.284 

1.654 

0.037* 

0.25 % 

-0.98 % 

-0.815 

0.299 

-0.088 

0.164 

-0.21 % 

-0.93 % 

-0.094 

0.48 

0.336 

0.145 

-0.52 % 

0.62 % 

4.747  <0.001 

1.839 

0.028* 

-0.58 % 

-1.19 % 

1.924 

0.09 

2.006 

0.02* 

-0.11 % 

-0.44 % 

3.386 

0.007* 

3.015 

0.004* 

-0.44 % 

-2.03 % 

1.991 

0.082 

3.648 

0.001* 

-0.27 % 

-0.1 % 

2.416 

0.043* 

2.019 

0.02* 

-0.28 % 

-0.31 % 

1.001 

0.249 

1.127 

0.075 

0.19 % 

-0.3 % 

0.635 

0.338 

0.869 

0.099 

0.07 % 

0.15 % 

0.900 

0.277 

0.735 

0.109 

-0.21 % 

-0.93 % 

1.709 

0.11 

1.659 

0.037* 

-0.2 % 

0.5 % 

0.619 

0.338 

0.020 

0.171 

-0.6 % 

-0.22 % 

3.498 

0.007* 

2.217 

0.015* 

-0.69 % 

-0.05 % 

1.805 

0.107 

1.092 

0.078 

 
 
Table A.2 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

8.095e-04 ± 
2.509e-05 

8.088e-04 ± 
2.644e-05 

8.177e-04 ± 
2.850e-05 

8.010e-04 ± 
2.579e-05 

8.390e-04 ± 
4.054e-05 

8.269e-04 ± 
4.470e-05 

8.334e-04 ± 
3.315e-05 

8.233e-04 ± 
3.552e-05 

8.257e-04 ± 
2.980e-05 

7.981e-04 ± 
3.203e-05 

8.208e-04 ± 
2.865e-05 

7.933e-04 ± 
3.078e-05 

8.255e-04 ± 
2.786e-05 

8.317e-04 ± 
3.290e-05 

8.370e-04 ± 
3.299e-05 

8.186e-04 ± 
3.557e-05 

8.116e-04 ± 
2.571e-05 

8.327e-04 ± 
2.444e-05 

8.057e-04 
± 2.998e-
05 

8.044e-04 
± 2.507e-
05 

8.119e-04 
± 3.848e-
05 

7.946e-04 
± 2.784e-
05 

8.360e-04 
± 3.969e-
05 

8.247e-04 
± 3.885e-
05 

8.354e-04 
± 3.495e-
05 

8.233e-04 
± 3.270e-
05 

8.265e-04 
± 3.850e-
05 

7.993e-04 
± 3.469e-
05 

8.160e-04 
± 3.953e-
05 

7.874e-04 
± 2.872e-
05 

8.209e-04 
± 3.525e-
05 

8.253e-04 
± 3.271e-
05 

8.305e-04 
± 4.623e-
05 

8.100e-04 
± 3.750e-
05 

8.142e-04 
± 2.716e-
05 

8.318e-04 
± 2.798e-
05 

7.828e-04 
± 4.072e-
05 

7.893e-04 
± 4.033e-
05 

7.848e-04 
± 4.277e-
05 

7.778e-04 
± 4.761e-
05 

8.243e-04 
± 3.169e-
05 

8.267e-04 
± 3.792e-
05 

8.344e-04 
± 4.703e-
05 

8.221e-04 
± 4.550e-
05 

8.250e-04 
± 6.229e-
05 

8.069e-04 
± 6.816e-
05 

7.968e-04 
± 5.194e-
05 

7.833e-04 
± 4.619e-
05 

8.027e-04 
± 5.094e-
05 

8.185e-04 
± 5.609e-
05 

8.121e-04 
± 5.641e-
05 

8.064e-04 
± 7.571e-
05 

7.926e-04 
± 3.163e-
05 

8.186e-04 
± 4.089e-
05 

7.759e-04 ± 
4.752e-05 

7.747e-04 ± 
3.372e-05 

7.662e-04 ± 
3.664e-05 

7.692e-04 ± 
4.500e-05 

8.229e-04 ± 
6.250e-05 

8.218e-04 ± 
5.292e-05 

8.312e-04 ± 
7.984e-05 

8.210e-04 ± 
7.266e-05 

8.151e-04 ± 
9.048e-05 

8.105e-04 ± 
1.036e-04 

7.935e-04 ± 
8.025e-05 

7.813e-04 ± 
6.742e-05 

7.926e-04 ± 
5.755e-05 

8.013e-04 ± 
5.377e-05 

7.900e-04 ± 
6.052e-05 

7.926e-04 ± 
6.976e-05 

7.837e-04 ± 
2.922e-05 

8.018e-04 ± 
3.281e-05 

-0.47 % 

-0.88 % 

3.193 

0.011* 

3.073 

0.004* 

-0.55 % 

-1.86 % 

2.265 

0.05 

3.891 

<0.001 

-0.71 % 

-2.38 % 

3.571 

0.007* 

4.261 

<0.001 

-0.8 % 

-1.11 % 

2.562 

0.043* 

2.804 

0.005* 

-0.35 % 

-0.17 % 

1.301 

0.197 

1.026 

0.082 

-0.26 % 

-0.6 % 

0.009 

0.494 

0.247 

0.151 

0.24 % 

-0.39 % 

-0.098 

0.48 

0.311 

0.146 

0.01 % 

-0.13 % 

0.106 

0.48 

0.185 

0.157 

0.09 % 

-1.2 % 

0.065 

0.485 

0.753 

0.109 

0.15 % 

0.45 % 

-0.727 

0.317 

-0.708 

0.109 

-0.58 % 

-0.41 % 

2.400 

0.043* 

1.575 

0.039* 

-0.74 % 

-0.25 % 

1.005 

0.249 

0.536 

0.121 

-0.55 % 

-1.27 % 

2.329 

0.048* 

2.464 

0.01* 

-0.77 % 

-2.1 % 

1.181 

0.229 

2.239 

0.015* 

-0.78 % 

-2.72 % 

2.219 

0.053 

2.900 

0.004* 

-1.05 % 

-1.71 % 

0.905 

0.277 

1.338 

0.057 

0.33 % 

-1.12 % 

2.430 

0.043* 

3.921 

<0.001 

-0.11 % 

-2.05 % 

1.709 

0.11 

3.666 

0.001* 

Tables A.2- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 159 

 
 
 
 
 
 
 
Table A.3 Results of Post Hoc Tract Specific Comparisons of Diffusion Tensor Imaging Axial 
Diffusivity 

TractSeg 
Abbv. 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

mTBI 

Controls 

Percent Change 

 P1 

 P2 

P1 

 P2 

mTBI 

 Controls 

q-value 
P11 

t-
score 
P1 

t-
score 
P2 

q-value P21 

-0.74 % 

-0.83 % 

4.394  <0.001  4.168 

<0.001 

-0.5 % 

-1.2 % 

2.662  <0.001  4.269 

<0.001 

-0.61 % 

-1.74 % 

2.537  <0.001  3.395 

<0.001 

-0.85 % 

-1.58 % 

1.269 

0.005* 

2.053 

0.005* 

-0.65 % 

-2.19 % 

4.402  <0.001  5.742 

<0.001 

-0.95 % 

-3.37 % 

0.382 

0.013* 

1.853 

0.007* 

-0.19 % 

-1.54 % 

2.726  <0.001  4.337 

<0.001 

-0.06 % 

-1.16 % 

2.187 

0.001* 

2.981 

<0.001 

0.16 % 

0.33 % 

2.925  <0.001  1.763 

0.008* 

0.47 % 

0.36 % 

2.358  <0.001  1.631 

0.009* 

0.25 % 

0.19 % 

1.945 

0.002* 

1.724 

0.008* 

-0.15 % 

-0.08 % 

0.727 

0.009* 

0.799 

0.031* 

0.02 % 

-0.58 % 

2.342  <0.001  2.817 

<0.001 

0.11 % 

0.05 % 

2.335  <0.001  1.693 

0.008* 

1.007e-03 ± 2.573e-
05 

9.847e-04 ± 
2.329e-05 

1.000e-03 ± 2.132e-
05 

9.878e-04 ± 
2.429e-05 

1.073e-03 ± 4.054e-
05 

1.053e-03 ± 
3.877e-05 

1.068e-03 ± 3.867e-
05 

1.061e-03 ± 
5.729e-05 

1.146e-03 ± 3.747e-
05 

1.113e-03 ± 
3.338e-05 

1.187e-03 ± 6.398e-
05 

1.192e-03 ± 
7.785e-05 

1.107e-03 ± 3.109e-
05 

1.087e-03 ± 
3.036e-05 

1.129e-03 ± 4.402e-
05 

1.105e-03 ± 
3.968e-05 

1.122e-03 ± 3.542e-
05 

1.097e-03 ± 
3.661e-05 

1.126e-03 ± 4.085e-
05 

1.099e-03 ± 
4.358e-05 

1.086e-03 ± 3.002e-
05 

1.068e-03 ± 
2.105e-05 

1.114e-03 ± 4.317e-
05 

1.105e-03 ± 
4.256e-05 

1.097e-03 ± 2.547e-
05 

1.080e-03 ± 
2.395e-05 

1.032e-03 ± 2.771e-
05 

1.013e-03 ± 
3.399e-05 

1.015e-
03 ± 
2.342e-
05 

1.005e-
03 ± 
2.231e-
05 

1.079e-
03 ± 
3.541e-
05 

1.077e-
03 ± 
4.010e-
05 

1.153e-
03 ± 
3.106e-
05 

1.198e-
03 ± 
5.031e-
05 

1.109e-
03 ± 
2.756e-
05 

1.129e-
03 ± 
3.908e-
05 

1.121e-
03 ± 
2.533e-
05 

1.121e-
03 ± 
2.864e-
05 

1.083e-
03 ± 
2.908e-
05 

1.116e-
03 ± 
5.499e-
05 

1.097e-
03 ± 
2.436e-
05 

1.031e-
03 ± 
2.378e-
05 

9.765e-04 
± 2.670e-
05 

9.759e-04 
± 1.501e-
05 

1.035e-03 
± 3.740e-
05 

1.044e-03 
± 4.945e-
05 

1.089e-03 
± 2.505e-
05 

1.152e-03 
± 7.825e-
05 

1.070e-03 
± 2.602e-
05 

1.092e-03 
± 4.430e-
05 

1.101e-03 
± 6.714e-
05 

1.103e-03 
± 7.685e-
05 

1.070e-03 
± 4.447e-
05 

1.104e-03 
± 5.584e-
05 

1.074e-03 
± 3.965e-
05 

1.013e-03 
± 6.512e-
05 

 160 

 
 
 
 
 
 
Table A.3 (cont’d)  

TractSeg 
Abbv. 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

 P2 

mTBI 

 Controls 

t-
score 
P1 

q-
value 
P11 

t-
score 
P2 

q-value P21 

-0.2 % 

-0.98 % 

2.201  0.001*  2.837 

<0.001 

-0.26 % 

-0.57 % 

5.193  <0.001  4.021 

<0.001 

-0.4 % 

0.41 % 

4.071  <0.001  2.424 

0.002* 

-0.12 % 

-1.49 % 

3.961  <0.001  4.826 

<0.001 

-0.24 % 

-1.55 % 

3.083  <0.001  4.644 

<0.001 

-1.89 % 

-1.85 % 

0.309  0.014*  0.271 

0.053 

-1.4 % 

-3.03 % 

0.426  0.013*  0.852 

0.029* 

-0.97 % 

-0.84 % 

1.114  0.006*  1.598 

0.009* 

-0.93 % 

-2.38 % 

0.898  0.008*  2.070 

0.005* 

-0.26 % 

-0.99 % 

2.736  <0.001  3.603 

<0.001 

-0.39 % 

-0.98 % 

1.511  0.003*  2.536 

0.002* 

-0.99 % 

-1.85 % 

1.881  0.002*  2.753 

0.001* 

-0.75 % 

-1.68 % 

0.750  0.009*  1.562 

0.01* 

1.024e-03 ± 2.496e-
05 

1.009e-03 ± 
2.887e-05 

1.136e-03 ± 3.320e-
05 

1.098e-03 ± 
3.481e-05 

1.095e-03 ± 2.623e-
05 

1.068e-03 ± 
3.136e-05 

1.129e-03 ± 3.320e-
05 

1.095e-03 ± 
4.753e-05 

1.121e-03 ± 2.925e-
05 

1.095e-03 ± 
4.575e-05 

2.046e-03 ± 2.231e-
04 

2.066e-03 ± 
2.208e-04 

2.048e-03 ± 2.239e-
04 

2.052e-03 ± 
2.155e-04 

9.534e-04 ± 3.663e-
05 

9.448e-04 ± 
4.185e-05 

9.298e-04 ± 3.252e-
05 

9.265e-04 ± 
4.801e-05 

1.053e-03 ± 2.968e-
05 

1.035e-03 ± 
2.153e-05 

1.043e-03 ± 2.640e-
05 

1.033e-03 ± 
3.013e-05 

1.061e-03 ± 4.141e-
05 

1.049e-03 ± 
3.176e-05 

1.045e-03 ± 3.940e-
05 

1.043e-03 ± 
4.700e-05 

1.027e-
03 ± 
2.690e-
05 

1.139e-
03 ± 
2.442e-
05 

1.100e-
03 ± 
2.556e-
05 

1.131e-
03 ± 
2.590e-
05 

1.124e-
03 ± 
2.818e-
05 

2.085e-
03 ± 
2.114e-
04 

2.077e-
03 ± 
2.037e-
04 

9.628e-
04 ± 
5.867e-
05 

9.386e-
04 ± 
4.624e-
05 

1.056e-
03 ± 
2.808e-
05 

1.047e-
03 ± 
3.094e-
05 

1.072e-
03 ± 
4.333e-
05 

1.053e-
03 ± 
4.591e-
05 

9.993e-04 
± 4.918e-
05 

1.092e-03 
± 5.409e-
05 

1.072e-03 
± 5.302e-
05 

1.078e-03 
± 5.047e-
05 

1.078e-03 
± 4.495e-
05 

2.028e-03 
± 2.725e-
04 

1.990e-03 
± 3.089e-
04 

9.368e-04 
± 3.789e-
05 

9.044e-04 
± 7.191e-
05 

1.024e-03 
± 2.312e-
05 

1.023e-03 
± 3.073e-
05 

1.030e-03 
± 3.628e-
05 

1.025e-03 
± 6.098e-
05 

 161 

 
 
 
 
 
 
Table A.3 (cont’d)  

mTBI 

Controls 

Percent Change 

TractSeg Abbv. 

mTBI P1 

mTBI P2 

Controls 
P1 

Controls 
P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

q-value 
P21 

t-
score 
P2 

MCP 

9.671e-04 ± 4.922e-05 

MLF_left 

1.033e-03 ± 2.652e-05 

MLF_right 

1.029e-03 ± 2.723e-05 

OR_left 

1.102e-03 ± 3.927e-05 

OR_right 

1.094e-03 ± 4.333e-05 

POPT_left 

1.109e-03 ± 2.802e-05 

POPT_right 

1.100e-03 ± 2.963e-05 

SCP_left 

1.057e-03 ± 2.989e-05 

SCP_right 

1.044e-03 ± 2.503e-05 

SLF_III_left 

1.013e-03 ± 2.510e-05 

SLF_III_right 

1.006e-03 ± 2.348e-05 

SLF_II_left 

1.017e-03 ± 2.591e-05 

SLF_II_right 

1.007e-03 ± 2.589e-05 

SLF_I_left 

1.051e-03 ± 3.764e-05 

SLF_I_right 

1.041e-03 ± 3.308e-05 

STR_left 

1.071e-03 ± 2.795e-05 

STR_right 

1.052e-03 ± 2.962e-05 

ST_FO_left 

1.059e-03 ± 4.021e-05 

ST_FO_right 

1.073e-03 ± 3.768e-05 

-0.47 % 

1.33 % 

1.399 

0.004* 

0.184 

0.057 

-0.05 % 

-0.51 % 

1.818 

0.002* 

2.177 

0.004* 

0.01 % 

-0.8 % 

1.235 

0.005* 

2.194 

0.004* 

-0.45 % 

-0.9 % 

1.600 

0.003* 

1.939 

0.006* 

-0.28 % 

-1.09 % 

0.061 

0.017* 

0.758 

0.032* 

0.28 % 

-0.34 % 

1.755 

0.002* 

1.903 

0.006* 

0.11 % 

0.08 % 

1.808 

0.002* 

1.533 

0.01* 

-0.61 % 

-0.58 % 

4.535  <0.001 

4.136  <0.001 

-0.79 % 

-1.18 % 

3.009  <0.001 

3.991  <0.001 

-0.47 % 

-0.79 % 

2.957  <0.001 

2.866  <0.001 

-0.31 % 

-1.09 % 

2.229 

0.001* 

3.263  <0.001 

-0.29 % 

-0.74 % 

2.021 

0.002* 

2.245 

0.003* 

-0.29 % 

-1.06 % 

1.842 

0.002* 

2.892  <0.001 

0.32 % 

-1.01 % 

-0.353 

0.013* 

0.606 

0.038* 

-0.01 % 

-0.73 % 

0.445 

0.013* 

0.955 

0.026* 

-0.42 % 

0.09 % 

5.964  <0.001 

3.229  <0.001 

-0.64 % 

-1.03 % 

2.922  <0.001 

2.912  <0.001 

-0.47 % 

-0.7 % 

2.957  <0.001 

2.790  <0.001 

-0.6 % 

-2.05 % 

1.986 

0.002* 

3.522  <0.001 

9.626e-04 
± 3.853e-
05 

1.033e-03 
± 2.832e-
05 

1.029e-03 
± 2.783e-
05 

1.097e-03 
± 4.030e-
05 

1.091e-03 
± 3.987e-
05 

1.112e-03 
± 3.210e-
05 

1.101e-03 
± 3.086e-
05 

1.050e-03 
± 3.242e-
05 

1.035e-03 
± 2.552e-
05 

1.008e-03 
± 2.923e-
05 

1.003e-03 
± 2.646e-
05 

1.014e-03 
± 2.883e-
05 

1.004e-03 
± 2.526e-
05 

1.054e-03 
± 4.073e-
05 

1.041e-03 
± 3.473e-
05 

1.067e-03 
± 3.761e-
05 

1.045e-03 
± 3.229e-
05 

1.054e-03 
± 4.817e-
05 

1.067e-03 
± 3.943e-
05 

9.479e-04 
± 3.816e-
05 

1.020e-03 
± 2.389e-
05 

1.019e-03 
± 2.749e-
05 

1.084e-03 
± 3.617e-
05 

1.093e-03 
± 4.489e-
05 

1.094e-03 
± 3.411e-
05 

1.084e-03 
± 3.327e-
05 

1.016e-03 
± 3.352e-
05 

1.018e-03 
± 4.347e-
05 

9.897e-04 
± 3.604e-
05 

9.898e-04 
± 2.919e-
05 

1.001e-03 
± 3.330e-
05 

9.925e-04 
± 3.316e-
05 

1.056e-03 
± 7.126e-
05 

1.036e-03 
± 5.265e-
05 

1.018e-03 
± 3.962e-
05 

1.025e-03 
± 4.226e-
05 

1.025e-03 
± 4.049e-
05 

1.051e-03 
± 4.138e-
05 

9.606e-04 
± 4.000e-
05 

1.015e-03 
± 3.585e-
05 

1.011e-03 
± 3.368e-
05 

1.074e-03 
± 4.706e-
05 

1.081e-03 
± 6.077e-
05 

1.090e-03 
± 6.348e-
05 

1.085e-03 
± 5.636e-
05 

1.010e-03 
± 3.981e-
05 

1.006e-03 
± 2.895e-
05 

9.819e-04 
± 4.455e-
05 

9.790e-04 
± 2.266e-
05 

9.931e-04 
± 4.407e-
05 

9.820e-04 
± 3.410e-
05 

1.045e-03 
± 8.891e-
05 

1.029e-03 
± 7.461e-
05 

1.019e-03 
± 8.800e-
05 

1.014e-03 
± 5.503e-
05 

1.018e-03 
± 3.978e-
05 

1.029e-03 
± 2.899e-
05 

 162 

 
 
 
 
 
Table A.3 (cont’d)  

mTBI 

Controls 

Percent Change 

TractSeg Abbv. 

mTBI P1 

mTBI P2 

Controls 
P1 

Controls 
P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

q-value 
P21 

t-
score 
P2 

ST_OCC_left 

1.082e-03 ± 3.268e-05 

ST_OCC_right 

1.062e-03 ± 3.557e-05 

ST_PAR_left 

1.064e-03 ± 2.882e-05 

ST_PAR_right 

1.055e-03 ± 2.873e-05 

ST_POSTC_left 

1.081e-03 ± 3.349e-05 

ST_POSTC_right 

1.058e-03 ± 3.114e-05 

ST_PREC_left 

1.076e-03 ± 2.947e-05 

ST_PREC_right 

1.053e-03 ± 3.045e-05 

ST_PREF_left 

1.056e-03 ± 2.776e-05 

ST_PREF_right 

1.054e-03 ± 3.014e-05 

ST_PREM_left 

1.053e-03 ± 3.357e-05 

ST_PREM_right 

1.044e-03 ± 3.148e-05 

T_OCC_left 

1.103e-03 ± 4.013e-05 

T_OCC_right 

1.090e-03 ± 4.353e-05 

T_PAR_left 

1.075e-03 ± 3.197e-05 

T_PAR_right 

1.076e-03 ± 3.398e-05 

T_POSTC_left 

1.083e-03 ± 3.168e-05 

T_POSTC_right 

1.059e-03 ± 3.156e-05 

T_PREC_left 

1.083e-03 ± 2.815e-05 

-0.36 % 

-0.73 % 

2.965  <0.001 

3.212  <0.001 

-0.32 % 

-0.69 % 

1.188 

0.005* 

1.644 

0.009* 

0.2 % 

-0.31 % 

1.660 

0.003* 

1.768 

0.008* 

-0.05 % 

-0.05 % 

1.810 

0.002* 

1.600 

0.009* 

-0.23 % 

-0.78 % 

2.818  <0.001 

2.426 

0.002* 

-0.43 % 

0.22 % 

1.395 

0.004* 

0.584 

0.038* 

-0.51 % 

-0.4 % 

4.681  <0.001 

3.214  <0.001 

-0.72 % 

-0.23 % 

2.633  <0.001 

1.975 

0.006* 

-0.4 % 

-1.1 % 

3.951  <0.001 

4.013  <0.001 

-0.55 % 

-1.72 % 

2.805  <0.001 

4.510  <0.001 

-0.57 % 

-2.27 % 

4.323  <0.001 

5.080  <0.001 

-0.71 % 

-1.21 % 

3.238  <0.001 

3.756  <0.001 

-0.48 % 

-0.9 % 

1.499 

0.003* 

1.798 

0.008* 

-0.3 % 

-1.03 % 

0.086 

0.017* 

0.709 

0.034* 

0.24 % 

-0.44 % 

0.621 

0.011* 

1.053 

0.023* 

-0.02 % 

-0.25 % 

0.747 

0.009* 

0.866 

0.029* 

-0.02 % 

-0.85 % 

1.464 

0.004* 

1.705 

0.008* 

-0.04 % 

0.14 % 

0.151 

0.016* 

-0.026 

0.064 

-0.49 % 

-0.55 % 

3.804  <0.001 

2.679 

0.001* 

1.078e-03 
± 3.508e-
05 

1.058e-03 
± 3.316e-
05 

1.066e-03 
± 3.255e-
05 

1.055e-03 
± 2.928e-
05 

1.078e-03 
± 4.209e-
05 

1.053e-03 
± 3.115e-
05 

1.070e-03 
± 3.950e-
05 

1.046e-03 
± 2.879e-
05 

1.052e-03 
± 3.464e-
05 

1.048e-03 
± 3.014e-
05 

1.047e-03 
± 4.465e-
05 

1.036e-03 
± 3.357e-
05 

1.097e-03 
± 4.116e-
05 

1.086e-03 
± 3.999e-
05 

1.078e-03 
± 3.489e-
05 

1.076e-03 
± 3.320e-
05 

1.083e-03 
± 3.916e-
05 

1.059e-03 
± 3.222e-
05 

1.078e-03 
± 3.880e-
05 

1.055e-03 
± 2.302e-
05 

1.049e-03 
± 3.533e-
05 

1.050e-03 
± 2.912e-
05 

1.040e-03 
± 2.724e-
05 

1.052e-03 
± 4.335e-
05 

1.044e-03 
± 4.490e-
05 

1.033e-03 
± 3.791e-
05 

1.029e-03 
± 3.861e-
05 

1.022e-03 
± 3.595e-
05 

1.028e-03 
± 4.096e-
05 

1.010e-03 
± 3.695e-
05 

1.011e-03 
± 4.473e-
05 

1.085e-03 
± 3.587e-
05 

1.088e-03 
± 4.596e-
05 

1.069e-03 
± 3.948e-
05 

1.068e-03 
± 3.920e-
05 

1.067e-03 
± 5.405e-
05 

1.057e-03 
± 5.480e-
05 

1.047e-03 
± 4.469e-
05 

1.047e-03 
± 2.724e-
05 

1.042e-03 
± 3.978e-
05 

1.047e-03 
± 5.636e-
05 

1.040e-03 
± 4.490e-
05 

1.043e-03 
± 7.582e-
05 

1.046e-03 
± 7.164e-
05 

1.029e-03 
± 6.310e-
05 

1.026e-03 
± 5.209e-
05 

1.011e-03 
± 4.110e-
05 

1.010e-03 
± 3.036e-
05 

9.868e-04 
± 2.854e-
05 

9.990e-04 
± 4.066e-
05 

1.076e-03 
± 4.708e-
05 

1.077e-03 
± 6.103e-
05 

1.064e-03 
± 7.090e-
05 

1.066e-03 
± 6.324e-
05 

1.058e-03 
± 8.290e-
05 

1.059e-03 
± 9.134e-
05 

1.042e-03 
± 7.164e-
05 

 163 

 
 
 
 
 
Table A.3 (cont’d)  

mTBI 

Controls 

Percent Change 

TractSeg Abbv. 

mTBI P1 

mTBI P2 

Controls 
P1 

Controls 
P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

q-value 
P21 

t-
score 
P2 

T_PREC_right 

1.049e-03 ± 3.081e-05 

T_PREF_left 

1.076e-03 ± 2.889e-05 

T_PREF_right 

1.074e-03 ± 3.428e-05 

T_PREM_left 

1.087e-03 ± 3.444e-05 

T_PREM_right 

1.065e-03 ± 3.573e-05 

UF_left 

1.057e-03 ± 2.840e-05 

UF_right 

1.085e-03 ± 2.861e-05 

1.042e-03 
± 2.871e-
05 

1.071e-03 
± 3.634e-
05 

1.067e-03 
± 3.426e-
05 

1.079e-03 
± 4.639e-
05 

1.057e-03 
± 3.830e-
05 

1.056e-03 
± 2.995e-
05 

1.080e-03 
± 2.984e-
05 

1.031e-03 
± 4.103e-
05 

1.047e-03 
± 4.445e-
05 

1.055e-03 
± 5.484e-
05 

1.054e-03 
± 5.017e-
05 

1.047e-03 
± 6.767e-
05 

1.036e-03 
± 3.224e-
05 

1.067e-03 
± 4.207e-
05 

1.026e-03 
± 6.015e-
05 

1.033e-03 
± 4.993e-
05 

1.035e-03 
± 4.852e-
05 

1.029e-03 
± 5.125e-
05 

1.030e-03 
± 6.163e-
05 

1.020e-03 
± 1.933e-
05 

1.045e-03 
± 1.620e-
05 

-0.62 % 

-0.39 % 

1.915 

0.002* 

1.514 

0.011* 

-0.45 % 

-1.37 % 

3.101  <0.001 

3.479  <0.001 

-0.67 % 

-1.85 % 

1.677 

0.003* 

2.957  <0.001 

-0.69 % 

-2.4 % 

2.989  <0.001 

3.796  <0.001 

-0.79 % 

-1.6 % 

1.414 

0.004* 

2.115 

0.004* 

-0.12 % 

-1.52 % 

2.553  <0.001 

4.554  <0.001 

-0.49 % 

-2.11 % 

1.962 

0.002* 

4.528  <0.001 

Tables A.3- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 164 

 
 
 
 
 
 
 
 
 
 
 
 
 
Table A.4 Results of Post Hoc Tract Specific Comparisons of Diffusion Tensor Imaging Radial 
Diffusivity 

TractSeg Abbv 

P1 

P2 

P1 

P2 

mTBI  Control 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-
value 
P11 

t-score 
P2 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

asd 

6.926e-04 ± 
2.510e-05 

6.776e-04 ± 
2.346e-05 

7.217e-04 ± 
3.371e-05 

7.305e-04 ± 
3.776e-05 

7.607e-04 ± 
2.680e-05 

6.890e-04 ± 
4.217e-05 

7.017e-04 ± 
2.825e-05 

7.207e-04 ± 
3.389e-05 

7.022e-04 ± 
3.236e-05 

6.899e-04 ± 
3.043e-05 

6.840e-04 ± 
3.575e-05 

6.882e-04 ± 
5.660e-05 

6.962e-04 ± 
3.001e-05 

6.770e-04 ± 
2.722e-05 

6.694e-04 ± 
2.930e-05 

6.676e-04 ± 
2.487e-05 

6.358e-04 ± 
2.449e-05 

6.770e-04 ± 
2.380e-05 

6.729e-04 ± 
2.448e-05 

1.372e-03 ± 
2.615e-04 

1.352e-03 ± 
2.505e-04 

6.374e-04 ± 
5.814e-05 

6.372e-04 ± 
4.490e-05 

6.853e-04 ± 
2.739e-05 

6.818e-04 ± 
3.044e-05 

6.988e-04 ± 
3.708e-05 

6.943e-04 ± 
3.747e-05 

6.880e-04 ± 
2.491e-05 

6.754e-04 ± 
1.976e-05 

7.173e-04 ± 
3.941e-05 

7.230e-04 ± 
3.543e-05 

7.601e-04 ± 
4.472e-05 

6.885e-04 ± 
4.672e-05 

7.024e-04 ± 
3.106e-05 

7.247e-04 ± 
5.072e-05 

7.087e-04 ± 
3.842e-05 

7.003e-04 ± 
4.238e-05 

6.876e-04 ± 
2.890e-05 

6.893e-04 ± 
3.900e-05 

6.984e-04 ± 
2.789e-05 

6.744e-04 ± 
2.774e-05 

6.667e-04 ± 
2.393e-05 

6.660e-04 ± 
3.272e-05 

6.347e-04 ± 
2.548e-05 

6.760e-04 ± 
3.174e-05 

6.726e-04 ± 
2.664e-05 

1.362e-03 ± 
2.560e-04 

1.345e-03 ± 
2.628e-04 

6.330e-04 ± 
4.076e-05 

6.344e-04 ± 
2.976e-05 

6.851e-04 ± 
2.548e-05 

6.800e-04 ± 
2.425e-05 

6.941e-04 ± 
2.885e-05 

6.904e-04 ± 
2.834e-05 

6.646e-04 ± 
3.528e-05 

6.650e-04 ± 
3.511e-05 

6.835e-04 ± 
4.824e-05 

7.041e-04 ± 
5.440e-05 

7.000e-04 ± 
4.786e-05 

6.606e-04 ± 
5.164e-05 

6.794e-04 ± 
5.003e-05 

6.995e-04 ± 
7.412e-05 

7.047e-04 ± 
9.832e-05 

6.961e-04 ± 
1.081e-04 

6.892e-04 ± 
6.888e-05 

6.963e-04 ± 
7.274e-05 

6.913e-04 ± 
6.451e-05 

6.717e-04 ± 
7.710e-05 

6.604e-04 ± 
5.967e-05 

6.469e-04 ± 
7.265e-05 

6.331e-04 ± 
7.721e-05 

6.514e-04 ± 
7.116e-05 

6.495e-04 ± 
6.193e-05 

1.356e-03 ± 
3.019e-04 

1.299e-03 ± 
3.085e-04 

6.304e-04 ± 
7.152e-05 

6.203e-04 ± 
6.577e-05 

6.691e-04 ± 
4.612e-05 

6.683e-04 ± 
3.391e-05 

6.752e-04 ± 
4.890e-05 

6.771e-04 ± 
4.371e-05 

6.710e-04 ± 
3.590e-05 

6.708e-04 ± 
2.842e-05 

6.977e-04 ± 
4.747e-05 

7.189e-04 ± 
5.714e-05 

7.125e-04 ± 
4.266e-05 

6.783e-04 ± 
6.571e-05 

6.889e-04 ± 
4.400e-05 

7.052e-04 ± 
5.606e-05 

6.990e-04 ± 
5.347e-05 

6.906e-04 ± 
5.817e-05 

6.831e-04 ± 
4.045e-05 

6.858e-04 ± 
3.193e-05 

6.912e-04 ± 
4.055e-05 

6.662e-04 ± 
4.741e-05 

6.615e-04 ± 
4.739e-05 

6.507e-04 ± 
4.765e-05 

6.237e-04 ± 
3.943e-05 

6.628e-04 ± 
6.547e-05 

6.634e-04 ± 
5.898e-05 

1.362e-03 ± 
2.757e-04 

1.351e-03 ± 
2.675e-04 

6.390e-04 ± 
6.150e-05 

6.457e-04 ± 
6.380e-05 

6.683e-04 ± 
2.269e-05 

6.720e-04 ± 
2.238e-05 

6.762e-04 ± 
2.638e-05 

6.808e-04 ± 
3.137e-05 

 165 

q-
val
ue 
P21 

0.0
27* 

0.2
17 

0.0
3* 

0.2
02 

-0.66 % 

-0.96 % 

2.703 

0.077 

3.039 

-0.32 % 

-0.87 % 

0.960 

0.47 

1.559 

-0.61 % 

-2.03 % 

2.250 

0.158 

2.899 

-1.02 % 

-2.06 % 

0.945 

0.47 

1.674 

-0.08 % 

-1.75 % 

5.445  <0.00

1 

4.699  <0.
001 

-0.07 % 

-2.6 % 

0.774 

0.529 

2.071 

0.11 % 

-1.39 % 

1.376 

0.374 

2.285 

0.55 % 

-0.81 % 

1.362 

0.374 

1.588 

0.93 % 

0.82 % 

0.298 

0.709 

0.251 

1.51 % 

0.81 % 

-0.061  0.725 

0.240 

0.53 % 

0.89 % 

0.089 

0.725 

-0.137 

0.16 % 

1.53 % 

0.156 

0.725 

-0.519 

0.31 % 

0.01 % 

0.535 

0.626 

0.658 

-0.4 % 

0.83 % 

1.163 

0.414 

0.217 

-0.4 % 

-0.17 % 

0.807 

0.517 

0.645 

-0.24 % 

-0.6 % 

1.895 

0.204 

1.541 

-0.17 % 

1.5 % 

1.489 

0.354 

0.143 

-0.15 % 

-1.72 % 

1.369 

0.374 

2.040 

-0.04 % 

-2.1 % 

0.961 

0.47 

2.233 

-0.7 % 

-0.4 % 

0.128 

0.725 

0.076 

-0.51 % 

-3.85 % 

0.006 

0.728 

0.593 

-0.69 % 

-1.35 % 

-0.096  0.725 

0.191 

-0.43 % 

-3.93 % 

-0.594  0.621 

1.258 

-0.02 % 

0.11 % 

2.207 

0.158 

1.844 

-0.26 % 

-0.54 % 

1.173 

0.414 

1.567 

-0.66 % 

-0.15 % 

2.211 

0.158 

1.978 

-0.56 % 

-0.55 % 

1.277 

0.408 

1.470 

0.1
38 

0.1
04 

0.2
17 

0.6
74 

0.6
74 

0.6
74 

0.5
97 

0.5
46 

0.6
74 

0.5
46 

0.2
17 

0.6
74 

0.1
39 

0.1
08 

0.6
74 

0.5
58 

0.6
74 

0.3
08 

0.1
67 

0.2
17 

0.1
42 

0.2
4 

 
 
 
 
 
Table A.4 (cont’d)  

TractSeg Abbv 

P1 

P2 

P1 

P2 

mTBI  Control 

mTBI 

Controls 

Percent Change 

t-
score 
P1 

q-
value 
P11 

t-score 
P2 

q-
valu
e 
P21 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

6.041e-04 ± 
4.311e-05 

6.978e-04 ± 
2.931e-05 

6.892e-04 ± 
3.101e-05 

7.003e-04 ± 
4.076e-05 

6.915e-04 ± 
4.588e-05 

6.928e-04 ± 
2.568e-05 

6.742e-04 ± 
2.835e-05 

6.519e-04 ± 
2.732e-05 

6.451e-04 ± 
2.198e-05 

6.876e-04 ± 
2.547e-05 

6.713e-04 ± 
2.392e-05 

7.067e-04 ± 
3.277e-05 

6.984e-04 ± 
3.077e-05 

7.430e-04 ± 
4.054e-05 

7.220e-04 ± 
3.788e-05 

6.387e-04 ± 
2.448e-05 

6.227e-04 ± 
2.252e-05 

6.794e-04 ± 
3.460e-05 

6.794e-04 ± 
3.338e-05 

6.844e-04 ± 
3.375e-05 

6.743e-04 ± 
3.673e-05 

7.027e-04 ± 
2.979e-05 

6.805e-04 ± 
3.013e-05 

7.006e-04 ± 
2.844e-05 

6.683e-04 ± 
2.854e-05 

6.905e-04 ± 
2.735e-05 

6.658e-04 ± 
2.834e-05 

6.046e-04 ± 
3.286e-05 

6.968e-04 ± 
2.710e-05 

6.905e-04 ± 
2.741e-05 

6.990e-04 ± 
3.968e-05 

6.903e-04 ± 
3.964e-05 

6.964e-04 ± 
3.043e-05 

6.769e-04 ± 
2.844e-05 

6.483e-04 ± 
3.494e-05 

6.422e-04 ± 
2.353e-05 

6.873e-04 ± 
2.821e-05 

6.730e-04 ± 
2.385e-05 

7.016e-04 ± 
3.297e-05 

6.947e-04 ± 
2.901e-05 

7.445e-04 ± 
4.786e-05 

7.195e-04 ± 
4.255e-05 

6.349e-04 ± 
3.693e-05 

6.193e-04 ± 
2.865e-05 

6.807e-04 ± 
3.755e-05 

6.773e-04 ± 
3.279e-05 

6.831e-04 ± 
3.157e-05 

6.726e-04 ± 
3.110e-05 

7.040e-04 ± 
3.110e-05 

6.816e-04 ± 
2.781e-05 

6.993e-04 ± 
3.708e-05 

6.682e-04 ± 
3.251e-05 

6.859e-04 ± 
3.627e-05 

6.613e-04 ± 
2.734e-05 

6.147e-04 ± 
7.219e-05 

6.879e-04 ± 
4.632e-05 

6.870e-04 ± 
4.959e-05 

6.899e-04 ± 
7.430e-05 

6.903e-04 ± 
5.206e-05 

6.931e-04 ± 
7.727e-05 

6.745e-04 ± 
7.402e-05 

6.262e-04 ± 
7.020e-05 

6.256e-04 ± 
4.642e-05 

6.689e-04 ± 
5.911e-05 

6.716e-04 ± 
5.256e-05 

6.871e-04 ± 
5.678e-05 

6.848e-04 ± 
4.679e-05 

7.513e-04 ± 
1.027e-04 

7.184e-04 ± 
8.837e-05 

6.160e-04 ± 
1.065e-04 

6.044e-04 ± 
6.454e-05 

6.439e-04 ± 
6.515e-05 

6.472e-04 ± 
3.854e-05 

6.689e-04 ± 
5.584e-05 

6.658e-04 ± 
3.611e-05 

6.992e-04 ± 
6.829e-05 

6.785e-04 ± 
5.953e-05 

6.831e-04 ± 
7.874e-05 

6.713e-04 ± 
8.250e-05 

6.650e-04 ± 
7.016e-05 

6.551e-04 ± 
6.024e-05 

0.08 % 

1.98 % 

0.102  0.725 

-0.818  0.48
8 

-0.14 % 

-0.6 % 

0.643  0.598 

0.986 

0.19 % 

-0.36 % 

-0.032  0.725 

0.378 

-0.19 %  0.41 % 

1.165  0.414 

0.657 

-0.18 % 

-0.33 % 

-0.080  0.725 

0.002 

0.52 % 

-0.38 % 

-0.336  0.707 

0.258 

0.41 % 

0.37 % 

0.231  0.725 

0.209 

-0.55 % 

-1.47 %  1.689  0.253 

1.763 

-0.44 % 

-2.76 %  0.177  0.725 

1.992 

-0.05 % 

-1.07 %  1.149  0.414 

1.776 

0.25 % 

-0.38 % 

-0.352  0.707 

0.159 

-0.72 % 

-1.56 %  0.813  0.517 

1.313 

-0.53 % 

-1.33 %  0.442 

0.67 

1.049 

0.41
4 

0.65
7 

0.54
6 

0.68
3 

0.67
4 

0.67
4 

0.17
6 

0.14
2 

0.17
6 

0.67
4 

0.28
9 

0.38
8 

0.19 % 

-0.96 % 

-0.995  0.47 

-0.385  0.65
7 

-0.35 % 

-1.07 % 

-0.312  0.709 

0.070 

-0.6 % 

1.07 % 

3.417  0.019

1.144 

* 

-0.53 % 

-1.33 %  1.137  0.414 

1.367 

0.18 % 

-0.23 %  3.405  0.019

2.918 

* 

-0.31 % 

-2.01 %  1.868  0.204 

3.135 

-0.2 % 

0.41 % 

1.996  0.202 

1.324 

-0.25 % 

-0.01 %  0.856  0.508 

0.759 

0.18 % 

-0.3 % 

0.147  0.725 

0.406 

0.16 % 

0.3 % 

0.440 

0.67 

0.298 

-0.19 % 

-1.05 %  1.030  0.461 

1.182 

0.67
4 

0.35
1 

0.27
9 

0.03
* 

0.02
4* 

0.28
9 

0.51
6 

0.65
7 

0.67
4 

0.33
9 

-0.03 %  0.72 % 

0.186  0.725 

-0.236  0.67
4 

-0.66 % 

-0.07 %  2.697  0.077 

1.631 

-0.68 %  0.08 % 

1.268  0.408 

0.601 

0.21
1 

0.55
8 

6.028e-04 ± 
5.747e-05 

6.920e-04 ± 
3.567e-05 

6.895e-04 ± 
4.188e-05 

6.871e-04 ± 
3.076e-05 

6.925e-04 ± 
3.584e-05 

6.958e-04 ± 
4.768e-05 

6.720e-04 ± 
4.540e-05 

6.356e-04 ± 
5.143e-05 

6.433e-04 ± 
6.362e-05 

6.761e-04 ± 
5.829e-05 

6.741e-04 ± 
3.759e-05 

6.981e-04 ± 
4.919e-05 

6.940e-04 ± 
4.574e-05 

7.586e-04 ± 
9.004e-05 

7.262e-04 ± 
7.028e-05 

6.095e-04 ± 
4.429e-05 

6.126e-04 ± 
5.202e-05 

6.454e-04 ± 
3.381e-05 

6.604e-04 ± 
4.087e-05 

6.662e-04 ± 
2.020e-05 

6.658e-04 ± 
2.164e-05 

7.013e-04 ± 
4.253e-05 

6.765e-04 ± 
3.588e-05 

6.903e-04 ± 
5.255e-05 

6.665e-04 ± 
5.109e-05 

6.655e-04 ± 
4.600e-05 

6.546e-04 ± 
3.867e-05 

 166 

 
 
 
 
 
Table A.4 (cont’d)  

TractSeg Abbv 

P1 

P2 

P1 

P2 

mTBI  Control 

mTBI 

Controls 

Percent Change 

t-
score 
P1 

q-
value 
P11 

t-score 
P2 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

6.862e-04 ± 
2.550e-05 

6.861e-04 ± 
2.578e-05 

7.003e-04 ± 
2.757e-05 

6.798e-04 ± 
2.468e-05 

7.072e-04 ± 
4.198e-05 

6.955e-04 ± 
4.611e-05 

7.125e-04 ± 
3.498e-05 

6.970e-04 ± 
3.743e-05 

6.971e-04 ± 
3.117e-05 

6.677e-04 ± 
3.483e-05 

6.898e-04 ± 
3.095e-05 

6.655e-04 ± 
3.227e-05 

7.001e-04 ± 
2.893e-05 

7.107e-04 ± 
3.313e-05 

7.121e-04 ± 
3.405e-05 

6.954e-04 ± 
3.690e-05 

6.887e-04 ± 
2.637e-05 

6.826e-04 ± 
2.901e-05 

6.823e-04 ± 
2.420e-05 

6.946e-04 ± 
3.653e-05 

6.738e-04 ± 
2.700e-05 

7.054e-04 ± 
4.044e-05 

6.939e-04 ± 
3.958e-05 

7.142e-04 ± 
3.744e-05 

6.972e-04 ± 
3.505e-05 

6.983e-04 ± 
4.104e-05 

6.697e-04 ± 
3.996e-05 

6.852e-04 ± 
4.194e-05 

6.598e-04 ± 
3.168e-05 

6.957e-04 ± 
3.602e-05 

7.047e-04 ± 
3.333e-05 

7.060e-04 ± 
4.735e-05 

6.867e-04 ± 
3.860e-05 

6.933e-04 ± 
2.901e-05 

6.631e-04 ± 
4.479e-05 

6.701e-04 ± 
4.125e-05 

6.724e-04 ± 
4.717e-05 

6.611e-04 ± 
5.068e-05 

6.937e-04 ± 
3.080e-05 

6.959e-04 ± 
3.580e-05 

7.171e-04 ± 
5.163e-05 

6.990e-04 ± 
4.984e-05 

7.039e-04 ± 
6.754e-05 

6.816e-04 ± 
7.665e-05 

6.714e-04 ± 
5.676e-05 

6.596e-04 ± 
5.046e-05 

6.806e-04 ± 
5.524e-05 

7.004e-04 ± 
5.745e-05 

6.912e-04 ± 
6.029e-05 

6.861e-04 ± 
8.031e-05 

6.710e-04 ± 
3.417e-05 

6.584e-04 ± 
5.170e-05 

6.570e-04 ± 
3.744e-05 

6.559e-04 ± 
4.144e-05 

6.543e-04 ± 
4.936e-05 

6.965e-04 ± 
7.268e-05 

6.940e-04 ± 
5.270e-05 

7.146e-04 ± 
8.467e-05 

6.988e-04 ± 
7.777e-05 

6.936e-04 ± 
9.480e-05 

6.863e-04 ± 
1.103e-04 

6.694e-04 ± 
8.506e-05 

6.588e-04 ± 
7.162e-05 

6.725e-04 ± 
6.216e-05 

6.844e-04 ± 
5.691e-05 

6.707e-04 ± 
6.564e-05 

6.738e-04 ± 
7.435e-05 

6.656e-04 ± 
3.905e-05 

-0.53 % 

-0.71 %  2.633  0.078 

2.461 

-0.55 % 

-1.96 %  1.873  0.204 

3.277 

-0.81 % 

-2.45 %  2.975  0.053 

3.652 

-0.88 % 

-1.02 %  2.045  0.195 

2.104 

-0.25 % 

0.4 % 

1.162  0.414 

0.648 

-0.23 % 

-0.27 % 

-0.029  0.725 

0.24 % 

-0.35 % 

-0.406  0.682 

0.03 % 

-0.04 % 

-0.177  0.725 

0.18 % 

-1.46 % 

-0.572  0.625 

0.3 % 

0.68 % 

-1.046  0.461 

-0.010  0.68
3 

-0.025  0.68
3 

-0.120  0.67
4 

0.295 

0.67
4 
-0.962  0.41
6 

-0.66 % 

-0.31 %  1.687  0.253 

1.045 

-0.84 % 

-0.13 %  0.550  0.626 

0.087 

-0.63 % 

-1.19 %  1.884  0.204 

1.925 

-0.85 % 

-2.29 %  0.907  0.484 

1.830 

-0.85 % 

-2.97 %  1.776  0.234 

2.427 

-1.25 % 

-1.8 % 

0.654  0.598 

0.949 

0.67 % 

-0.81 %  2.174  0.158 

3.138 

q-
valu
e 
P21 

0.07
9 

0.02
4* 

0.01
2* 

0.13
7 

0.54
6 

0.38
8 

0.67
4 

0.15
1 

0.16
7 

0.07
9 

0.41
6 

0.02
4* 

0.03
1* 

UF_right 

7.064e-04 ± 
2.464e-05 

7.077e-04 ± 
3.061e-05 

6.943e-04 ± 
4.186e-05 
Tables A.4- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

6.804e-04 ± 
4.518e-05 

-2.01 %  1.448  0.364 

0.19 % 

2.848 

 167 

 
 
 
 
 
 
 
 
Table A.5 Results of Post Hoc Tract Specific Comparisons of Diffusion Kurtosis Imaging 
Fractional Anisotropy 

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

0.200 ± 0.010 

0.200 ± 0.008  0.202 ± 0.014  0.204 ± 0.013 

0.02 % 

0.77 % 

0.204 ± 0.008 

0.203 ± 0.008  0.202 ± 0.011  0.201 ± 0.020 

-0.53 % 

-0.71 % 

0.204 ± 0.008 

0.204 ± 0.009  0.210 ± 0.015  0.213 ± 0.014 

0.26 % 

0.199 ± 0.008 

0.200 ± 0.009  0.203 ± 0.013  0.204 ± 0.016 

0.27 % 

1.53 % 

0.77 % 

0.208 ± 0.013 

0.207 ± 0.018  0.220 ± 0.016  0.219 ± 0.027 

-0.34 % 

-0.44 % 

0.258 ± 0.016 

0.255 ± 0.023  0.264 ± 0.017  0.257 ± 0.041 

-0.98 % 

-2.65 % 

0.224 ± 0.010 

0.223 ± 0.010  0.227 ± 0.016  0.228 ± 0.021 

-0.34 % 

0.224 ± 0.012 

0.223 ± 0.017  0.229 ± 0.025  0.231 ± 0.028 

-0.67 % 

0.234 ± 0.013 

0.232 ± 0.014  0.235 ± 0.024  0.238 ± 0.025 

-0.64 % 

0.243 ± 0.013 

0.241 ± 0.015  0.243 ± 0.025  0.245 ± 0.029 

-0.98 % 

0.234 ± 0.013 

0.234 ± 0.010  0.230 ± 0.017  0.229 ± 0.021 

-0.27 % 

0.236 ± 0.012 

0.235 ± 0.012  0.233 ± 0.009  0.229 ± 0.030 

-0.33 % 

0.228 ± 0.010 

0.228 ± 0.009  0.228 ± 0.016  0.228 ± 0.021 

-0.3 % 

0.215 ± 0.012 

0.217 ± 0.014  0.217 ± 0.018  0.216 ± 0.022 

0.95 % 

0.219 ± 0.013 

0.219 ± 0.012  0.218 ± 0.017  0.217 ± 0.019 

0.21 % 

0.255 ± 0.010 

0.257 ± 0.011  0.260 ± 0.019  0.264 ± 0.018 

0.45 % 

0.262 ± 0.010 

0.262 ± 0.011  0.263 ± 0.018  0.263 ± 0.024 

0.07 % 

0.246 ± 0.010 

0.247 ± 0.011  0.249 ± 0.020  0.252 ± 0.020 

0.44 % 

0.43 % 

1.07 % 

1.09 % 

1.06 % 

-0.14 % 

-1.92 % 

0.06 % 

-0.37 % 

-0.16 % 

1.47 % 

0.09 % 

1.2 % 

0.248 ± 0.010 

0.248 ± 0.010  0.249 ± 0.017  0.251 ± 0.020 

0.15 % 

1 % 

0.185 ± 0.037 

0.186 ± 0.039  0.192 ± 0.038  0.191 ± 0.046 

0.55 % 

-0.54 % 

0.189 ± 0.037 

0.190 ± 0.042  0.193 ± 0.039  0.198 ± 0.042 

0.56 % 

2.1 % 

0.200 ± 0.020 

0.200 ± 0.022  0.189 ± 0.023  0.189 ± 0.031 

-0.15 % 

-0.38 % 

0.189 ± 0.013 

0.187 ± 0.014  0.178 ± 0.019  0.181 ± 0.024 

-1.15 % 

1.79 % 

0.217 ± 0.009 

0.216 ± 0.009  0.218 ± 0.010  0.215 ± 0.023 

-0.26 % 

-1.35 % 

0.216 ± 0.008 

0.215 ± 0.009  0.216 ± 0.008  0.215 ± 0.021 

-0.37 % 

-0.7 % 

0.214 ± 0.009 

0.213 ± 0.011  0.217 ± 0.007  0.212 ± 0.027 

-0.45 % 

-2.37 % 

0.214 ± 0.009 

0.213 ± 0.013  0.216 ± 0.009  0.212 ± 0.028 

-0.29 % 

-1.6 % 

0.218 ± 0.012 

0.217 ± 0.012  0.212 ± 0.023  0.209 ± 0.026 

-0.53 % 

-1.08 % 

0.207 ± 0.011 

0.208 ± 0.008  0.204 ± 0.014  0.206 ± 0.013 

0.33 % 

0.85 % 

0.214 ± 0.010 

0.213 ± 0.009  0.209 ± 0.015  0.208 ± 0.019 

-0.46 % 

-0.05 % 

0.223 ± 0.011 

0.223 ± 0.011  0.223 ± 0.007  0.220 ± 0.028 

-0.08 % 

0.231 ± 0.011 

0.231 ± 0.011  0.230 ± 0.009  0.228 ± 0.020 

-0.09 % 

0.238 ± 0.009 

0.238 ± 0.010  0.236 ± 0.017  0.239 ± 0.017 

0.12 % 

0.249 ± 0.010 

0.249 ± 0.010  0.248 ± 0.017  0.250 ± 0.020 

-0.06 % 

0.230 ± 0.013 

0.231 ± 0.017  0.229 ± 0.019  0.233 ± 0.023 

0.4 % 

0.232 ± 0.009 

0.231 ± 0.012  0.226 ± 0.020  0.231 ± 0.021 

-0.32 % 

0.205 ± 0.011 

0.204 ± 0.010  0.205 ± 0.021  0.208 ± 0.017 

-0.62 % 

-1.31 % 

-0.66 % 

1.2 % 

0.78 % 

1.81 % 

2.19 % 

1.43 % 

0.209 ± 0.009 

0.206 ± 0.011  0.203 ± 0.014  0.200 ± 0.028 

-1.33 % 

-1.53 % 

0.193 ± 0.011 

0.194 ± 0.011  0.192 ± 0.015  0.198 ± 0.018 

0.69 % 

0.194 ± 0.010 

0.194 ± 0.010  0.191 ± 0.014  0.194 ± 0.015 

0.08 % 

0.193 ± 0.010 

0.194 ± 0.012  0.190 ± 0.022  0.193 ± 0.018 

0.44 % 

0.201 ± 0.011 

0.203 ± 0.012  0.199 ± 0.019  0.203 ± 0.017 

0.58 % 

0.247 ± 0.011 

0.249 ± 0.014  0.254 ± 0.019  0.256 ± 0.026 

0.53 % 

0.255 ± 0.012 

0.255 ± 0.012  0.257 ± 0.020  0.261 ± 0.018 

0.01 % 

2.71 % 

1.46 % 

1.79 % 

2.23 % 

1.05 % 

1.7 % 

0.225 ± 0.012 

0.223 ± 0.013  0.233 ± 0.013  0.231 ± 0.026 

-0.77 % 

-0.72 % 

0.228 ± 0.013 

0.227 ± 0.016  0.232 ± 0.014  0.229 ± 0.028 

-0.52 % 

-1.02 % 

0.225 ± 0.011 

0.225 ± 0.011  0.225 ± 0.010  0.221 ± 0.028 

0 % 

-1.4 % 

-0.689 

0.654 

-2.185 

-1.386 

-3.209 

-1.265 

-0.810 

-0.953 

-0.235 

0.086 

1.209 

0.855 

0.182 

-0.410 

0.308 

-1.391 

-0.231 

-1.067 

-0.244 

-0.669 

-0.378 

1.800 

2.748 

-0.416 

-0.049 

-1.156 

-0.850 

1.495 

0.860 

1.560 

-0.034 

0.352 

0.627 

0.375 

0.302 

1.696 

0.048 

2.040 

0.067 

0.909 

0.788 

0.737 

-1.720 

-0.474 

-2.225 

-0.991 

0.241 

0.638 

0.638 

0.365 

0.525 

0.111 

0.615 

0.638 

0.638 

0.722 

0.752 

0.638 

0.638 

0.722 

0.722 

0.722 

0.525 

0.722 

0.638 

0.722 

0.638 

0.722 

0.375 

0.211 

0.722 

0.752 

0.638 

0.638 

0.485 

0.638 

0.457 

0.752 

0.722 

0.645 

0.722 

0.722 

0.375 

0.752 

0.365 

0.752 

0.638 

0.638 

0.638 

0.375 

0.709 

0.365 

0.638 

0.722 

-1.403 

0.605 

-3.155 

-1.555 

-2.250 

-0.188 

-1.203 

-1.491 

-1.075 

-0.879 

1.262 

1.373 

-0.065 

0.273 

0.542 

-2.030 

-0.206 

-1.490 

-0.823 

-0.453 

-0.602 

1.643 

1.356 

0.359 

0.174 

0.302 

0.153 

1.632 

0.692 

1.251 

0.587 

0.628 

-0.143 

-0.226 

-0.384 

0.095 

-1.184 

1.363 

-1.045 

0.058 

0.128 

-0.156 

-1.624 

-1.622 

-1.721 

-0.503 

0.873 

0.383 

0.665 

0.125 

0.383 

0.245 

0.738 

0.402 

0.383 

0.457 

0.551 

0.383 

0.383 

0.738 

0.738 

0.671 

0.308 

0.738 

0.383 

0.566 

0.684 

0.665 

0.383 

0.383 

0.727 

0.738 

0.738 

0.738 

0.383 

0.656 

0.383 

0.665 

0.665 

0.738 

0.738 

0.722 

0.738 

0.402 

0.383 

0.457 

0.738 

0.738 

0.738 

0.383 

0.383 

0.383 

0.679 

0.551 

 168 

 
 
 
 
 
 
 
Table A.5 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

0.229 ± 
0.011 

0.217 ± 
0.010 

0.228 ± 
0.010 

0.222 ± 
0.010 

0.234 ± 
0.011 

0.222 ± 
0.009 

0.230 ± 
0.009 

0.215 ± 
0.009 

0.214 ± 
0.008 

0.207 ± 
0.009 

0.216 ± 
0.008 

0.220 ± 
0.011 

0.227 ± 
0.011 

0.216 ± 
0.010 

0.228 ± 
0.011 

0.227 ± 
0.010 

0.239 ± 
0.013 

0.225 ± 
0.009 

0.230 ± 
0.010 

0.215 ± 
0.009 

0.209 ± 
0.008 

0.212 ± 
0.009 

0.217 ± 
0.010 

0.221 ± 
0.012 

0.215 ± 
0.010 

0.228 ± 
0.011 

0.218 ± 
0.009 

0.228 ± 
0.009 

0.223 ± 
0.011 

0.233 ± 
0.012 

0.223 ± 
0.011 

0.230 ± 
0.009 

0.216 ± 
0.008 

0.214 ± 
0.008 

0.208 ± 
0.009 

0.216 ± 
0.009 

0.220 ± 
0.011 

0.227 ± 
0.011 

0.217 ± 
0.009 

0.228 ± 
0.010 

0.228 ± 
0.012 

0.238 ± 
0.015 

0.226 ± 
0.012 

0.231 ± 
0.011 

0.216 ± 
0.010 

0.210 ± 
0.009 

0.214 ± 
0.011 

0.219 ± 
0.011 

0.219 ± 
0.014 

0.214 ± 
0.015 

0.227 ± 
0.010 

0.215 ± 
0.016 

0.226 ± 
0.014 

0.226 ± 
0.017 

0.235 ± 
0.017 

0.229 ± 
0.016 

0.231 ± 
0.015 

0.220 ± 
0.015 

0.217 ± 
0.012 

0.213 ± 
0.017 

0.218 ± 
0.016 

0.221 ± 
0.006 

0.226 ± 
0.009 

0.214 ± 
0.016 

0.226 ± 
0.015 

0.226 ± 
0.021 

0.235 ± 
0.025 

0.231 ± 
0.019 

0.231 ± 
0.018 

0.220 ± 
0.016 

0.212 ± 
0.013 

0.219 ± 
0.017 

0.220 ± 
0.020 

0.224 ± 
0.014 

0.218 ± 
0.011 

0.226 ± 
0.020 

0.217 ± 
0.015 

0.228 ± 
0.016 

0.230 ± 
0.015 

0.238 ± 
0.019 

0.231 ± 
0.014 

0.234 ± 
0.016 

0.221 ± 
0.016 

0.218 ± 
0.016 

0.215 ± 
0.015 

0.221 ± 
0.018 

0.218 ± 
0.027 

0.224 ± 
0.019 

0.217 ± 
0.017 

0.228 ± 
0.016 

0.233 ± 
0.019 

0.238 ± 
0.024 

0.235 ± 
0.017 

0.234 ± 
0.016 

0.223 ± 
0.017 

0.214 ± 
0.015 

0.224 ± 
0.016 

0.224 ± 
0.019 

0.220 ± 
0.026 

0.216 ± 
0.028 

-0.34 % 

-0.46 % 

0.496 

0.706 

0.476 

0.679 

0.42 % 

1.09 % 

0.555 

0.688 

0.126 

0.738 

-0.25 % 

0.76 % 

0.760 

0.638 

0.026 

0.738 

0.57 % 

1.82 % 

-1.133 

0.638 

-1.970 

0.313 

-0.22 % 

1.24 % 

-0.199 

0.722 

-1.058 

0.457 

0.53 % 

1.18 % 

-2.122 

0.365 

-2.506 

0.245 

0.18 % 

1.25 % 

-0.515 

0.706 

-1.305 

0.383 

0.29 % 

0.41 % 

-1.718 

0.375 

-1.817 

0.383 

0.03 % 

0.69 % 

-1.011 

0.638 

-1.445 

0.383 

0.46 % 

1.27 % 

-1.720 

0.375 

-2.409 

0.245 

0.26 % 

1.1 % 

-0.803 

0.638 

-1.298 

0.383 

-0.02 % 

-1.35 % 

-0.193 

0.722 

0.548 

0.671 

-0.04 % 

-0.68 % 

0.299 

0.722 

0.638 

0.665 

0.36 % 

1.36 % 

0.657 

0.638 

-0.006 

0.739 

0.08 % 

1.05 % 

0.695 

0.638 

0.042 

0.738 

0.32 % 

2.76 % 

0.141 

0.738 

-1.270 

0.383 

-0.28 % 

1.38 % 

0.756 

0.638 

-0.077 

0.738 

0.56 % 

1.79 % 

-1.735 

0.375 

-2.291 

0.245 

0.45 % 

1.47 % 

-0.190 

0.722 

-0.862 

0.551 

0.49 % 

1.05 % 

-1.813 

0.375 

-2.121 

0.285 

0.39 % 

1.22 % 

-0.985 

0.638 

-1.527 

0.383 

0.66 % 

2.24 % 

-2.032 

0.365 

-2.885 

0.138 

0.72 % 

1.63 % 

-0.818 

0.638 

-1.332 

0.383 

-1.12 % 

-1.77 % 

-0.787 

0.638 

-0.292 

0.738 

-0.82 % 

-0.92 % 

-0.913 

0.638 

-0.479 

0.679 

Tables A.5- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 169 

 
 
 
 
 
 
Table A.6 Results of Post Hoc Tract Specific Comparisons of Diffusion Kurtosis Imaging Mean 
Diffusivity 

TractSeg 
Abbv. 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

9.405e-04 ± 
2.940e-05 

9.296e-04 ± 
2.842e-05 

9.992e-04 ± 
4.547e-05 

1.011e-03 ± 
5.214e-05 

1.051e-03 ± 
3.639e-05 

1.027e-03 ± 
5.777e-05 

9.983e-04 ± 
3.397e-05 

1.023e-03 ± 
4.521e-05 

1.015e-03 ± 
3.549e-05 

1.004e-03 ± 
3.478e-05 

9.667e-04 ± 
4.284e-05 

9.930e-04 ± 
7.692e-05 

9.882e-04 ± 
3.531e-05 

9.343e-04 ± 
2.881e-05 

9.228e-04 ± 
3.196e-05 

1.005e-03 ± 
3.049e-05 

9.612e-04 ± 
3.090e-05 

1.005e-03 ± 
2.968e-05 

9.983e-04 ± 
3.213e-05 

2.084e-03 ± 
3.622e-04 

2.055e-03 ± 
3.465e-04 

8.986e-04 ± 
8.235e-05 

8.806e-04 ± 
6.453e-05 

9.595e-04 ± 
3.503e-05 

9.506e-04 ± 
4.005e-05 

9.724e-04 ± 
5.191e-05 

9.614e-04 ± 
5.392e-05 

9.323e-04 ± 
3.288e-05 

9.242e-04 ± 
2.501e-05 

9.917e-04 ± 
5.552e-05 

9.990e-04 ± 
5.000e-05 

1.045e-03 ± 
5.522e-05 

1.019e-03 ± 
6.614e-05 

9.973e-04 ± 
4.112e-05 

1.026e-03 ± 
6.617e-05 

1.021e-03 ± 
4.994e-05 

1.015e-03 ± 
5.405e-05 

9.703e-04 ± 
3.623e-05 

9.924e-04 ± 
5.292e-05 

9.895e-04 ± 
3.479e-05 

9.320e-04 ± 
2.868e-05 

9.186e-04 ± 
2.781e-05 

1.002e-03 ± 
4.362e-05 

9.573e-04 ± 
3.144e-05 

1.003e-03 ± 
4.230e-05 

9.962e-04 ± 
3.487e-05 

2.053e-03 ± 
3.602e-04 

2.035e-03 ± 
3.660e-04 

8.888e-04 ± 
5.192e-05 

8.727e-04 ± 
4.092e-05 

9.568e-04 ± 
3.508e-05 

9.457e-04 ± 
3.215e-05 

9.619e-04 ± 
4.314e-05 

9.529e-04 ± 
4.118e-05 

9.049e-04 ± 
3.749e-05 

9.134e-04 ± 
3.055e-05 

9.651e-04 ± 
5.828e-05 

9.922e-04 ± 
7.752e-05 

9.882e-04 ± 
4.902e-05 

1.018e-03 ± 
9.343e-05 

9.747e-04 ± 
4.772e-05 

9.953e-04 ± 
6.128e-05 

9.975e-04 ± 
5.900e-05 

9.915e-04 ± 
6.630e-05 

9.569e-04 ± 
3.907e-05 

9.851e-04 ± 
4.702e-05 

9.744e-04 ± 
4.187e-05 

9.146e-04 ± 
5.173e-05 

9.050e-04 ± 
4.795e-05 

9.677e-04 ± 
5.499e-05 

9.326e-04 ± 
4.490e-05 

9.738e-04 ± 
7.679e-05 

9.747e-04 ± 
7.129e-05 

2.071e-03 ± 
3.822e-04 

2.050e-03 ± 
3.706e-04 

8.881e-04 ± 
6.794e-05 

8.802e-04 ± 
7.328e-05 

9.338e-04 ± 
2.693e-05 

9.348e-04 ± 
3.207e-05 

9.423e-04 ± 
3.590e-05 

9.451e-04 ± 
4.685e-05 

8.943e-04 ± 
4.150e-05 

9.025e-04 ± 
3.455e-05 

9.416e-04 ± 
6.102e-05 

9.710e-04 ± 
7.387e-05 

9.673e-04 ± 
5.410e-05 

9.792e-04 ± 
6.515e-05 

9.563e-04 ± 
5.529e-05 

9.825e-04 ± 
8.887e-05 

1.003e-03 ± 
1.205e-04 

9.959e-04 ± 
1.314e-04 

9.626e-04 ± 
7.973e-05 

9.946e-04 ± 
8.659e-05 

9.700e-04 ± 
7.486e-05 

9.169e-04 ± 
9.080e-05 

8.982e-04 ± 
7.526e-05 

9.620e-04 ± 
9.214e-05 

9.427e-04 ± 
9.243e-05 

9.531e-04 ± 
8.568e-05 

9.528e-04 ± 
7.834e-05 

2.052e-03 ± 
4.225e-04 

1.971e-03 ± 
4.431e-04 

8.751e-04 ± 
8.569e-05 

8.455e-04 ± 
9.221e-05 

9.277e-04 ± 
4.550e-05 

9.272e-04 ± 
3.808e-05 

9.315e-04 ± 
5.478e-05 

9.341e-04 ± 
6.133e-05 

 170 

-0.87 % 

-1.18 % 

3.941 

0.002* 

3.874  <0.001 

-0.58 % 

-1.19 % 

1.948 

0.066 

2.831 

0.004* 

-0.75 % 

-2.43 % 

2.438 

0.031* 

3.133 

0.002* 

-1.2 % 

-2.13 % 

1.126 

0.197 

1.780 

0.025* 

-0.58 % 

-2.11 % 

5.509  <0.001 

4.988  <0.001 

-0.81 % 

-3.77 % 

0.484 

0.348 

2.118 

0.015* 

-0.1 % 

-1.89 % 

2.194 

0.047* 

3.272 

0.002* 

0.35 % 

-1.29 % 

1.934 

0.066 

2.174 

0.013* 

0.61 % 

0.6 % 

1.475 

0.139 

0.895 

0.076 

1.13 % 

0.44 % 

0.973 

0.229 

0.881 

0.076 

0.37 % 

0.59 % 

0.797 

0.275 

0.563 

0.102 

-0.06 % 

0.96 % 

0.378 

0.377 

-0.127 

0.141 

0.13 % 

-0.45 % 

1.303 

0.17 

1.503 

0.037* 

-0.25 % 

0.24 % 

1.966 

0.066 

1.107 

0.065 

-0.46 % 

-0.75 % 

1.720 

0.091 

1.716 

0.027* 

-0.34 % 

-0.59 % 

3.510 

0.004* 

2.458 

0.008* 

-0.4 % 

1.08 % 

2.888 

0.017* 

1.031 

0.071 

-0.18 % 

-2.13 % 

2.504 

0.03* 

3.292 

0.002* 

-0.21 % 

-2.25 % 

1.900 

0.069 

3.263 

0.002* 

-1.49 % 

-0.92 % 

0.119 

0.461 

0.006 

0.151 

-1.01 % 

-3.89 % 

0.049 

0.461 

0.594 

0.1 

-1.08 % 

-1.47 % 

0.453 

0.354 

0.809 

0.081 

-0.9 % 

-3.94 % 

0.018 

0.461 

1.735 

0.026* 

-0.28 % 

-0.66 % 

2.635 

0.027* 

2.761 

0.004* 

-0.51 % 

-0.82 % 

1.407 

0.153 

1.964 

0.018* 

-1.07 % 

-1.15 % 

2.108 

0.054 

2.357 

0.01* 

-0.88 % 

-1.16 % 

1.069 

0.211 

1.451 

0.038* 

 
 
 
 
 
 
 
Table A.6 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

8.708e-04 ± 
6.393e-05 

9.510e-04 ± 
3.528e-05 

9.415e-04 ± 
3.877e-05 

9.945e-04 ± 
5.449e-05 

9.831e-04 ± 
6.108e-05 

9.991e-04 ± 
3.297e-05 

9.794e-04 ± 
3.744e-05 

9.493e-04 ± 
3.634e-05 

9.361e-04 ± 
2.950e-05 

9.434e-04 ± 
3.034e-05 

9.267e-04 ± 
2.895e-05 

9.559e-04 ± 
3.748e-05 

9.449e-04 ± 
3.611e-05 

9.992e-04 ± 
5.026e-05 

9.798e-04 ± 
4.602e-05 

9.481e-04 ± 
2.894e-05 

9.235e-04 ± 
2.917e-05 

9.471e-04 ± 
4.708e-05 

9.540e-04 ± 
4.446e-05 

9.709e-04 ± 
4.304e-05 

9.519e-04 ± 
4.755e-05 

9.781e-04 ± 
3.716e-05 

9.555e-04 ± 
3.860e-05 

9.937e-04 ± 
3.872e-05 

9.563e-04 ± 
3.757e-05 

9.860e-04 ± 
3.607e-05 

9.556e-04 ± 
3.735e-05 

8.685e-04 ± 
4.820e-05 

9.494e-04 ± 
3.444e-05 

8.572e-04 ± 
6.320e-05 

9.384e-04 ± 
3.773e-05 

8.784e-04 ± 
8.677e-05 

9.319e-04 ± 
5.611e-05 

9.415e-04 ± 
3.484e-05 

9.359e-04 ± 
4.410e-05 

9.289e-04 ± 
5.616e-05 

9.898e-04 ± 
5.405e-05 

9.795e-04 ± 
5.414e-05 

1.003e-03 ± 
3.879e-05 

9.816e-04 ± 
3.672e-05 

9.421e-04 ± 
4.372e-05 

9.290e-04 ± 
3.024e-05 

9.396e-04 ± 
3.632e-05 

9.254e-04 ± 
2.677e-05 

9.506e-04 ± 
3.974e-05 

9.399e-04 ± 
3.350e-05 

1.002e-03 ± 
5.790e-05 

9.771e-04 ± 
5.018e-05 

9.744e-04 ± 
4.405e-05 

9.841e-04 ± 
5.234e-05 

9.932e-04 ± 
5.520e-05 

9.688e-04 ± 
5.193e-05 

9.138e-04 ± 
5.534e-05 

9.207e-04 ± 
7.218e-05 

9.186e-04 ± 
6.050e-05 

9.196e-04 ± 
3.960e-05 

9.367e-04 ± 
5.163e-05 

9.310e-04 ± 
4.893e-05 

1.011e-03 ± 
1.104e-04 

9.788e-04 ± 
8.196e-05 

9.733e-04 ± 
8.667e-05 

9.772e-04 ± 
7.327e-05 

9.883e-04 ± 
9.957e-05 

9.708e-04 ± 
9.200e-05 

9.047e-04 ± 
7.949e-05 

9.000e-04 ± 
5.096e-05 

9.067e-04 ± 
6.975e-05 

9.110e-04 ± 
5.053e-05 

9.242e-04 ± 
6.602e-05 

9.181e-04 ± 
5.575e-05 

9.988e-04 ± 
1.340e-04 

9.684e-04 ± 
1.147e-04 

-0.27 % 

2.47 % 

0.735 

0.284 

-0.608 

0.099 

-0.18 % 

-0.69 % 

1.220 

0.174 

1.550 

0.035* 

0 % 

-0.75 % 

0.483 

0.348 

1.111 

0.065 

-0.47 % 

-0.12 % 

1.315 

0.17 

0.943 

0.076 

-0.37 % 

-0.7 % 

-0.058 

0.461 

0.138 

0.141 

0.39 % 

-0.49 % 

0.528 

0.342 

0.912 

0.076 

0.22 % 

0.2 % 

0.898 

0.246 

0.725 

0.088 

-0.75 % 

-0.99 % 

3.000 

0.014* 

2.499 

0.008* 

-0.75 % 

-2.25 % 

1.273 

0.174 

2.899 

0.003* 

-0.4 % 

-1.29 % 

2.241 

0.045* 

2.579 

0.007* 

-0.15 % 

-0.94 % 

0.784 

0.275 

1.544 

0.035* 

-0.56 % 

-1.33 % 

1.630 

0.106 

2.013 

0.017* 

-0.54 % 

-1.39 % 

1.237 

0.174 

1.979 

0.018* 

0.29 % 

-1.24 % 

-0.630 

0.313 

0.147 

0.141 

-0.27 % 

-1.06 % 

0.064 

0.461 

0.450 

0.112 

9.418e-04 ± 
4.814e-05 

8.943e-04 ± 
5.409e-05 

9.001e-04 ± 
1.305e-04 

-0.66 % 

0.65 % 

5.250 

<0.001 

2.022 

0.017* 

9.162e-04 ± 
3.585e-05 

9.451e-04 ± 
5.533e-05 

9.478e-04 ± 
4.387e-05 

9.666e-04 ± 
4.266e-05 

9.476e-04 ± 
4.183e-05 

9.795e-04 ± 
3.927e-05 

9.552e-04 ± 
3.533e-05 

9.907e-04 ± 
5.073e-05 

9.534e-04 ± 
3.951e-05 

9.786e-04 ± 
5.040e-05 

8.993e-04 ± 
6.302e-05 

9.020e-04 ± 
4.637e-05 

9.270e-04 ± 
5.471e-05 

9.416e-04 ± 
2.495e-05 

9.384e-04 ± 
3.310e-05 

9.692e-04 ± 
4.808e-05 

9.441e-04 ± 
4.045e-05 

9.698e-04 ± 
6.816e-05 

9.484e-04 ± 
6.615e-05 

9.413e-04 ± 
5.738e-05 

8.863e-04 ± 
8.140e-05 

8.959e-04 ± 
6.989e-05 

9.033e-04 ± 
3.635e-05 

9.403e-04 ± 
5.582e-05 

9.355e-04 ± 
4.555e-05 

9.649e-04 ± 
8.628e-05 

9.445e-04 ± 
7.279e-05 

9.574e-04 ± 
1.078e-04 

9.513e-04 ± 
1.081e-04 

9.381e-04 ± 
9.289e-05 

-0.8 % 

-1.45 % 

2.186 

0.047* 

2.173 

0.013* 

-0.21 % 

-0.67 % 

3.320 

0.006* 

2.980 

0.003* 

-0.65 % 

-2.56 % 

2.002 

0.066 

3.723 

<0.001 

-0.44 % 

-0.14 % 

2.514 

0.03* 

2.046 

0.017* 

-0.45 % 

-0.3 % 

1.034 

0.218 

1.006 

0.073 

0.14 % 

-0.45 % 

0.775 

0.275 

0.991 

0.073 

-0.03 % 

0.04 % 

1.009 

0.222 

0.835 

0.079 

-0.3 % 

-1.27 % 

1.795 

0.082 

1.771 

0.025* 

-0.31 % 

0.31 % 

0.617 

0.313 

0.125 

0.141 

-0.75 % 

-0.34 % 

3.749 

0.002* 

2.337 

0.01* 

9.469e-04 ± 
3.475e-05 

9.327e-04 ± 
5.116e-05 

9.306e-04 ± 
7.729e-05 

-0.91 % 

-0.22 % 

1.958 

0.066 

1.234 

0.055 

 171 

 
 
 
 
 
Table A.6 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

9.640e-04 ± 
3.279e-05 

9.622e-04 ± 
3.418e-05 

9.727e-04 ± 
3.862e-05 

9.570e-04 ± 
3.336e-05 

1.001e-03 ± 
5.633e-05 

9.842e-04 ± 
6.157e-05 

9.924e-04 ± 
4.415e-05 

9.798e-04 ± 
4.823e-05 

9.904e-04 ± 
3.971e-05 

9.554e-04 ± 
4.229e-05 

9.895e-04 ± 
3.827e-05 

9.530e-04 ± 
4.062e-05 

9.875e-04 ± 
3.675e-05 

9.944e-04 ± 
4.383e-05 

9.990e-04 ± 
4.493e-05 

9.810e-04 ± 
4.743e-05 

9.459e-04 ± 
3.425e-05 

9.729e-04 ± 
3.170e-05 

9.582e-04 ± 
4.179e-05 

9.550e-04 ± 
3.441e-05 

9.648e-04 ± 
5.527e-05 

9.475e-04 ± 
3.796e-05 

9.958e-04 ± 
5.548e-05 

9.801e-04 ± 
5.434e-05 

9.945e-04 ± 
4.689e-05 

9.789e-04 ± 
4.424e-05 

9.908e-04 ± 
5.222e-05 

9.565e-04 ± 
4.577e-05 

9.262e-04 ± 
5.308e-05 

9.342e-04 ± 
5.387e-05 

9.254e-04 ± 
5.624e-05 

9.252e-04 ± 
6.306e-05 

9.813e-04 ± 
4.437e-05 

9.846e-04 ± 
5.332e-05 

9.928e-04 ± 
6.218e-05 

9.773e-04 ± 
6.115e-05 

9.876e-04 ± 
8.681e-05 

9.661e-04 ± 
9.372e-05 

9.136e-04 ± 
6.012e-05 

9.133e-04 ± 
4.582e-05 

8.975e-04 ± 
4.967e-05 

9.101e-04 ± 
5.958e-05 

9.800e-04 ± 
8.538e-05 

9.783e-04 ± 
7.429e-05 

9.876e-04 ± 
1.092e-04 

9.748e-04 ± 
9.990e-05 

9.722e-04 ± 
1.256e-04 

9.680e-04 ± 
1.417e-04 

9.824e-04 ± 
5.481e-05 

9.537e-04 ± 
6.981e-05 

9.490e-04 ± 
1.109e-04 

9.440e-04 ± 
3.812e-05 

9.809e-04 ± 
4.894e-05 

9.846e-04 ± 
4.471e-05 

9.901e-04 ± 
6.565e-05 

9.687e-04 ± 
5.084e-05 

9.485e-04 ± 
3.537e-05 

9.369e-04 ± 
6.181e-05 

9.550e-04 ± 
6.698e-05 

9.749e-04 ± 
7.562e-05 

9.630e-04 ± 
7.541e-05 

9.640e-04 ± 
1.017e-04 

9.204e-04 ± 
4.060e-05 

9.331e-04 ± 
9.125e-05 

9.383e-04 ± 
7.597e-05 

9.510e-04 ± 
7.470e-05 

9.316e-04 ± 
8.453e-05 

9.429e-04 ± 
9.568e-05 

9.080e-04 ± 
4.121e-05 

9.701e-04 ± 
3.521e-05 

9.535e-04 ± 
5.527e-05 

9.301e-04 ± 
4.556e-05 

-0.6 % 

-1.36 % 

3.463 

0.004* 

3.428 

0.001* 

-0.75 % 

-2.24 % 

2.490 

0.03* 

3.995 

<0.001 

-0.81 % 

-3.02 % 

3.824 

0.002* 

4.409 

<0.001 

-0.99 % 

-1.64 % 

2.682 

0.026* 

3.071 

0.003* 

-0.53 % 

-0.14 % 

1.259 

0.174 

0.896 

0.076 

-0.41 % 

-0.63 % 

-0.023 

0.461 

0.110 

0.141 

0.21 % 

-0.53 % 

-0.031 

0.461 

0.376 

0.119 

-0.1 % 

-0.26 % 

0.170 

0.448 

0.242 

0.134 

0.04 % 

-1.56 % 

0.183 

0.448 

0.894 

0.076 

0.12 % 

0.19 % 

-0.659 

0.308 

-0.536 

0.103 

-0.72 % 

-0.49 % 

2.674 

0.026* 

1.689 

0.028* 

-0.94 % 

-0.41 % 

1.216 

0.174 

0.723 

0.088 

-0.67 % 

-1.76 % 

2.525 

0.03* 

2.720 

0.005* 

-0.98 % 

-2.44 % 

1.303 

0.17 

2.276 

0.011* 

-0.9 % 

-3.26 % 

2.378 

0.034* 

2.965 

0.003* 

-1.26 % 

-2.19 % 

0.947 

0.233 

1.462 

0.038* 

0.28 % 

-1.35 % 

2.471 

0.03* 

3.918 

<0.001 

-0.29 % 

-2.46 % 

1.786 

0.082 

3.777 

<0.001 

Tables A.6- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 172 

 
 
 
 
 
 
 
 
 
 
Table A.7 Results of Post Hoc Tract Specific Comparisons of Diffusion Kurtosis Imaging Axial 
Diffusivity 

TractSeg 
Abbv. 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI  

 Controls 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P2 

-0.86 % 

-1.15 % 

4.927  <0.001 

4.659  <0.001 

-0.68 % 

-1.42 % 

2.574 

0.008* 

4.230  <0.001 

-0.7 % 

-2.24 % 

2.755 

0.006* 

3.578  <0.001 

-1.1 % 

-2.09 % 

1.395 

0.057 

2.279 

0.005* 

-0.69 % 

-2.44 % 

5.236  <0.001 

6.108  <0.001 

-0.85 % 

-4.21 % 

0.371 

0.196 

2.264 

0.005* 

-0.25 % 

-1.97 % 

2.868 

0.004* 

4.425  <0.001 

0.02 % 

-1.48 % 

2.387 

0.011* 

2.958 

0.001* 

0.23 % 

0.19 % 

2.770 

0.006* 

1.702 

0.014* 

0.63 % 

0.08 % 

2.117 

0.019* 

1.634 

0.015* 

0.22 % 

0.21 % 

1.733 

0.036* 

1.526 

0.017* 

-0.13 % 

0.18 % 

0.655 

0.15 

0.575 

0.06 

-0.02 % 

-0.75 % 

2.240 

0.015* 

2.644 

0.002* 

-0.06 % 

-0.15 % 

2.516 

0.009* 

2.012 

0.008* 

-0.39 % 

-1.08 % 

2.479 

0.009* 

2.772 

0.002* 

-0.4 % 

-0.69 % 

5.110  <0.001 

3.610  <0.001 

-0.48 % 

0.46 % 

4.096  <0.001 

2.246 

0.005* 

-0.23 % 

-1.98 % 

3.733  <0.001 

4.548  <0.001 

-0.29 % 

-2.09 % 

2.951 

0.004* 

4.517  <0.001 

-1.57 % 

-1.56 % 

0.261 

0.215 

0.252 

0.083 

-1.11 % 

-3.82 % 

0.241 

0.216 

0.878 

0.045* 

-1.1 % 

-1.47 % 

1.017 

0.096 

1.641 

0.015* 

-1.08 % 

-3.33 % 

0.647 

0.15 

2.171 

0.006* 

-0.31 % 

-1.11 % 

2.968 

0.004* 

3.738  <0.001 

-0.54 % 

-1.11 % 

1.593 

0.043* 

2.526 

0.003* 

-1.1 % 

-1.82 % 

2.110 

0.019* 

2.837 

0.001* 

-0.88 % 

-1.66 % 

0.966 

0.102 

1.647 

0.015* 

1.137e-03 ± 
2.870e-05 

1.126e-03 ± 
2.878e-05 

1.207e-03 ± 
4.550e-05 

1.216e-03 ± 
5.206e-05 

1.285e-03 ± 
3.787e-05 

1.318e-03 ± 
6.172e-05 

1.238e-03 ± 
3.436e-05 

1.267e-03 ± 
4.850e-05 

1.264e-03 ± 
3.244e-05 

1.260e-03 ± 
3.440e-05 

1.204e-03 ± 
3.913e-05 

1.240e-03 ± 
7.382e-05 

1.225e-03 ± 
3.237e-05 

1.148e-03 ± 
3.020e-05 

1.136e-03 ± 
3.228e-05 

1.278e-03 ± 
3.066e-05 

1.232e-03 ± 
3.136e-05 

1.265e-03 ± 
3.123e-05 

1.260e-03 ± 
3.421e-05 

2.431e-03 ± 
3.183e-04 

2.408e-03 ± 
3.038e-04 

1.081e-03 ± 
8.088e-05 

1.051e-03 ± 
6.341e-05 

1.181e-03 ± 
3.524e-05 

1.168e-03 ± 
4.034e-05 

1.195e-03 ± 
5.468e-05 

1.179e-03 ± 
5.903e-05 

1.127e-03 ± 
3.362e-05 

1.119e-03 ± 
2.558e-05 

1.199e-03 ± 
5.503e-05 

1.203e-03 ± 
4.988e-05 

1.276e-03 ± 
5.025e-05 

1.306e-03 ± 
7.864e-05 

1.235e-03 ± 
4.059e-05 

1.267e-03 ± 
6.164e-05 

1.267e-03 ± 
4.850e-05 

1.268e-03 ± 
5.256e-05 

1.207e-03 ± 
3.687e-05 

1.239e-03 ± 
5.439e-05 

1.225e-03 ± 
3.325e-05 

1.147e-03 ± 
2.997e-05 

1.132e-03 ± 
2.951e-05 

1.272e-03 ± 
4.403e-05 

1.226e-03 ± 
3.289e-05 

1.263e-03 ± 
4.285e-05 

1.256e-03 ± 
3.649e-05 

2.393e-03 ± 
3.217e-04 

2.381e-03 ± 
3.238e-04 

1.069e-03 ± 
5.278e-05 

1.040e-03 ± 
4.193e-05 

1.177e-03 ± 
3.770e-05 

1.162e-03 ± 
3.385e-05 

1.181e-03 ± 
5.091e-05 

1.169e-03 ± 
4.907e-05 

1.095e-03 ± 
3.274e-05 

1.105e-03 ± 
3.145e-05 

1.169e-03 ± 
5.415e-05 

1.193e-03 ± 
7.617e-05 

1.223e-03 ± 
4.878e-05 

1.310e-03 ± 
9.903e-05 

1.207e-03 ± 
4.365e-05 

1.232e-03 ± 
5.686e-05 

1.234e-03 ± 
5.206e-05 

1.235e-03 ± 
5.934e-05 

1.185e-03 ± 
3.034e-05 

1.227e-03 ± 
5.163e-05 

1.204e-03 ± 
3.478e-05 

1.123e-03 ± 
4.682e-05 

1.112e-03 ± 
3.971e-05 

1.226e-03 ± 
4.775e-05 

1.192e-03 ± 
4.279e-05 

1.221e-03 ± 
6.699e-05 

1.224e-03 ± 
6.438e-05 

2.407e-03 ± 
3.339e-04 

2.386e-03 ± 
3.260e-04 

1.059e-03 ± 
5.528e-05 

1.039e-03 ± 
6.510e-05 

1.152e-03 ± 
2.899e-05 

1.150e-03 ± 
3.763e-05 

1.163e-03 ± 
4.061e-05 

1.163e-03 ± 
5.609e-05 

1.082e-03 ± 
3.656e-05 

1.089e-03 ± 
2.307e-05 

1.143e-03 ± 
5.492e-05 

1.168e-03 ± 
6.844e-05 

1.194e-03 ± 
3.581e-05 

1.255e-03 ± 
8.869e-05 

1.184e-03 ± 
4.207e-05 

1.214e-03 ± 
7.173e-05 

1.237e-03 ± 
1.013e-04 

1.236e-03 ± 
1.116e-04 

1.188e-03 ± 
6.570e-05 

1.229e-03 ± 
7.370e-05 

1.195e-03 ± 
6.046e-05 

1.121e-03 ± 
8.185e-05 

1.100e-03 ± 
6.918e-05 

1.218e-03 ± 
8.127e-05 

1.197e-03 ± 
7.733e-05 

1.197e-03 ± 
7.479e-05 

1.199e-03 ± 
6.964e-05 

2.369e-03 ± 
3.880e-04 

2.295e-03 ± 
4.230e-04 

1.043e-03 ± 
6.917e-05 

1.004e-03 ± 
9.658e-05 

1.139e-03 ± 
3.157e-05 

1.137e-03 ± 
3.810e-05 

1.142e-03 ± 
4.550e-05 

1.143e-03 ± 
7.127e-05 

 173 

 
 
 
 
 
 
Table A.7 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  

 Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

1.071e-03 ± 
6.466e-05 

1.155e-03 ± 
3.379e-05 

1.148e-03 ± 
3.761e-05 

1.229e-03 ± 
5.251e-05 

1.221e-03 ± 
5.889e-05 

1.244e-03 ± 
3.416e-05 

1.231e-03 ± 
3.839e-05 

1.179e-03 ± 
3.726e-05 

1.164e-03 ± 
3.064e-05 

1.143e-03 ± 
3.106e-05 

1.128e-03 ± 
2.968e-05 

1.143e-03 ± 
3.460e-05 

1.131e-03 ± 
3.491e-05 

1.190e-03 ± 
4.919e-05 

1.176e-03 ± 
4.418e-05 

1.201e-03 ± 
3.187e-05 

1.177e-03 ± 
3.361e-05 

1.179e-03 ± 
5.018e-05 

1.192e-03 ± 
4.641e-05 

1.206e-03 ± 
4.186e-05 

1.184e-03 ± 
4.732e-05 

1.196e-03 ± 
3.686e-05 

1.180e-03 ± 
3.836e-05 

1.221e-03 ± 
4.183e-05 

1.188e-03 ± 
3.885e-05 

1.214e-03 ± 
3.774e-05 

1.185e-03 ± 
3.839e-05 

1.066e-03 ± 
5.086e-05 

1.153e-03 ± 
3.526e-05 

1.146e-03 ± 
3.513e-05 

1.223e-03 ± 
5.395e-05 

1.216e-03 ± 
5.357e-05 

1.247e-03 ± 
3.929e-05 

1.231e-03 ± 
3.888e-05 

1.170e-03 ± 
4.364e-05 

1.154e-03 ± 
3.182e-05 

1.136e-03 ± 
3.644e-05 

1.123e-03 ± 
2.680e-05 

1.138e-03 ± 
3.768e-05 

1.126e-03 ± 
3.241e-05 

1.192e-03 ± 
5.473e-05 

1.173e-03 ± 
4.698e-05 

1.194e-03 ± 
4.787e-05 

1.168e-03 ± 
3.931e-05 

1.175e-03 ± 
6.145e-05 

1.184e-03 ± 
5.045e-05 

1.201e-03 ± 
4.476e-05 

1.179e-03 ± 
4.336e-05 

1.197e-03 ± 
3.992e-05 

1.178e-03 ± 
3.605e-05 

1.217e-03 ± 
5.251e-05 

1.183e-03 ± 
3.930e-05 

1.206e-03 ± 
5.141e-05 

1.175e-03 ± 
3.579e-05 

1.047e-03 ± 
5.328e-05 

1.136e-03 ± 
3.268e-05 

1.136e-03 ± 
3.777e-05 

1.205e-03 ± 
4.734e-05 

1.219e-03 ± 
5.695e-05 

1.227e-03 ± 
4.715e-05 

1.212e-03 ± 
4.598e-05 

1.130e-03 ± 
4.542e-05 

1.134e-03 ± 
6.193e-05 

1.110e-03 ± 
4.811e-05 

1.111e-03 ± 
3.722e-05 

1.118e-03 ± 
4.439e-05 

1.111e-03 ± 
4.475e-05 

1.194e-03 ± 
9.878e-05 

1.168e-03 ± 
7.384e-05 

1.135e-03 ± 
5.223e-05 

1.142e-03 ± 
5.843e-05 

1.132e-03 ± 
5.240e-05 

1.162e-03 ± 
5.716e-05 

1.171e-03 ± 
2.893e-05 

1.167e-03 ± 
4.299e-05 

1.179e-03 ± 
3.994e-05 

1.161e-03 ± 
3.700e-05 

1.186e-03 ± 
6.182e-05 

1.172e-03 ± 
6.255e-05 

1.159e-03 ± 
5.239e-05 

1.155e-03 ± 
5.168e-05 

1.066e-03 ± 
6.899e-05 

1.128e-03 ± 
5.063e-05 

1.125e-03 ± 
4.764e-05 

1.196e-03 ± 
6.880e-05 

1.206e-03 ± 
7.741e-05 

1.220e-03 ± 
9.075e-05 

1.212e-03 ± 
8.075e-05 

1.120e-03 ± 
6.357e-05 

1.115e-03 ± 
4.157e-05 

1.097e-03 ± 
6.196e-05 

1.096e-03 ± 
3.389e-05 

1.107e-03 ± 
6.121e-05 

1.098e-03 ± 
4.853e-05 

1.180e-03 ± 
1.248e-04 

1.158e-03 ± 
1.053e-04 

1.137e-03 ± 
1.172e-04 

1.127e-03 ± 
7.560e-05 

1.118e-03 ± 
5.352e-05 

1.130e-03 ± 
3.634e-05 

1.163e-03 ± 
3.766e-05 

1.160e-03 ± 
4.903e-05 

1.173e-03 ± 
7.820e-05 

1.160e-03 ± 
6.385e-05 

1.172e-03 ± 
1.026e-04 

1.173e-03 ± 
9.963e-05 

1.153e-03 ± 
8.677e-05 

1.150e-03 ± 
7.155e-05 

 174 

-0.44 % 

1.75 % 

1.289 

0.067 

0.022 

0.099 

-0.15 % 

-0.69 % 

1.943 

0.025* 

2.276 

0.005* 

-0.14 % 

-0.91 % 

1.138 

0.082 

1.965 

0.009* 

-0.5 % 

-0.72 % 

1.651 

0.04* 

1.686 

0.014* 

-0.36 % 

-1.06 % 

0.093 

0.237 

0.600 

0.06 

0.23 % 

-0.55 % 

1.582 

0.043* 

1.745 

0.013* 

0.07 % 

0 % 

1.645 

0.04* 

1.401 

0.02* 

-0.75 % 

-0.85 % 

4.365  <0.001 

3.674  <0.001 

-0.85 % 

-1.74 % 

2.622 

0.007* 

4.094  <0.001 

-0.58 % 

-1.18 % 

3.262 

0.003* 

3.264  <0.001 

-0.43 % 

-1.37 % 

1.812 

0.032* 

3.354  <0.001 

-0.44 % 

-0.95 % 

2.354 

0.012* 

2.496 

0.003* 

-0.5 % 

-1.22 % 

1.882 

0.028* 

2.750 

0.002* 

0.23 % 

-1.18 % 

-0.229 

0.216 

0.600 

0.06 

-0.23 % 

-0.89 % 

0.500 

0.173 

0.845 

0.046* 

-0.63 % 

0.16 % 

6.210  <0.001 

2.956 

0.001* 

-0.83 % 

-1.34 % 

3.051 

0.003* 

2.937 

0.001* 

-0.35 % 

-1.2 % 

3.211 

0.003* 

3.350  <0.001 

-0.69 % 

-2.8 % 

2.120 

0.019* 

4.018  <0.001 

-0.41 % 

-0.71 % 

3.062 

0.003* 

3.150  <0.001 

-0.48 % 

-0.62 % 

1.293 

0.067 

1.511 

0.017* 

0.12 % 

-0.51 % 

1.586 

0.043* 

1.740 

0.013* 

-0.15 % 

-0.1 % 

1.676 

0.04* 

1.468 

0.018* 

-0.33 % 

-1.19 % 

2.598 

0.008* 

2.424 

0.003* 

-0.43 % 

0.1 % 

1.201 

0.076 

0.575 

0.06 

-0.72 % 

-0.55 % 

4.642  <0.001 

3.100  <0.001 

-0.87 % 

-0.37 % 

2.581 

0.008* 

1.938 

0.009* 

 
 
 
 
 
Table A.7 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  

 Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P2 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

1.183e-03 ± 
3.438e-05 

1.180e-03 ± 
3.657e-05 

1.182e-03 ± 
4.221e-05 

1.175e-03 ± 
3.808e-05 

1.233e-03 ± 
5.412e-05 

1.217e-03 ± 
5.928e-05 

1.210e-03 ± 
4.214e-05 

1.205e-03 ± 
4.604e-05 

1.222e-03 ± 
4.028e-05 

1.190e-03 ± 
4.058e-05 

1.221e-03 ± 
3.738e-05 

1.182e-03 ± 
3.977e-05 

1.208e-03 ± 
3.669e-05 

1.209e-03 ± 
4.391e-05 

1.217e-03 ± 
4.501e-05 

1.202e-03 ± 
4.657e-05 

1.173e-03 ± 
3.496e-05 

1.201e-03 ± 
3.376e-05 

1.176e-03 ± 
4.390e-05 

1.172e-03 ± 
3.706e-05 

1.173e-03 ± 
5.867e-05 

1.164e-03 ± 
4.155e-05 

1.226e-03 ± 
5.529e-05 

1.212e-03 ± 
5.384e-05 

1.211e-03 ± 
4.516e-05 

1.203e-03 ± 
4.326e-05 

1.221e-03 ± 
5.075e-05 

1.189e-03 ± 
4.220e-05 

1.213e-03 ± 
5.310e-05 

1.172e-03 ± 
3.713e-05 

1.200e-03 ± 
4.826e-05 

1.198e-03 ± 
4.404e-05 

1.207e-03 ± 
6.484e-05 

1.189e-03 ± 
5.014e-05 

1.173e-03 ± 
3.626e-05 

1.196e-03 ± 
3.558e-05 

1.138e-03 ± 
4.938e-05 

1.146e-03 ± 
5.420e-05 

1.126e-03 ± 
5.127e-05 

1.135e-03 ± 
6.074e-05 

1.209e-03 ± 
4.722e-05 

1.215e-03 ± 
5.855e-05 

1.202e-03 ± 
5.434e-05 

1.196e-03 ± 
5.554e-05 

1.205e-03 ± 
7.790e-05 

1.190e-03 ± 
8.110e-05 

1.175e-03 ± 
6.239e-05 

1.157e-03 ± 
5.724e-05 

1.168e-03 ± 
6.137e-05 

1.183e-03 ± 
7.357e-05 

1.173e-03 ± 
6.971e-05 

1.177e-03 ± 
9.451e-05 

1.144e-03 ± 
4.304e-05 

1.178e-03 ± 
5.722e-05 

1.122e-03 ± 
5.255e-05 

1.121e-03 ± 
4.259e-05 

1.094e-03 ± 
4.357e-05 

1.117e-03 ± 
5.415e-05 

1.200e-03 ± 
6.843e-05 

1.203e-03 ± 
7.828e-05 

1.195e-03 ± 
9.944e-05 

1.192e-03 ± 
8.996e-05 

1.189e-03 ± 
1.154e-04 

1.190e-03 ± 
1.282e-04 

1.167e-03 ± 
1.014e-04 

1.151e-03 ± 
8.315e-05 

1.148e-03 ± 
6.800e-05 

1.156e-03 ± 
6.869e-05 

1.139e-03 ± 
7.681e-05 

1.153e-03 ± 
8.559e-05 

1.123e-03 ± 
2.854e-05 

1.147e-03 ± 
2.801e-05 

-0.56 % 

-1.46 % 

4.070  <0.001 

4.228  <0.001 

-0.73 % 

-2.19 % 

2.903 

0.004* 

4.699  <0.001 

-0.75 % 

-2.84 % 

4.344  <0.001 

4.994  <0.001 

-0.9 % 

-1.59 % 

3.133 

0.003* 

3.748  <0.001 

-0.54 % 

-0.73 % 

1.548 

0.045* 

1.575 

0.016* 

-0.39 % 

-0.99 % 

0.119 

0.235 

0.554 

0.061 

0.15 % 

-0.58 % 

0.552 

0.165 

0.936 

0.042* 

-0.16 % 

-0.33 % 

0.677 

0.149 

0.729 

0.053 

-0.08 % 

-1.33 % 

1.133 

0.082 

1.612 

0.015* 

-0.07 % 

0.01 % 

0.013 

0.25 

-0.041 

0.098 

-0.7 % 

-0.65 % 

3.710  <0.001 

2.461 

0.003* 

-0.86 % 

-0.51 % 

1.955 

0.025* 

1.457 

0.018* 

-0.61 % 

-1.74 % 

3.179 

0.003* 

3.502  <0.001 

-0.9 % 

-2.3 % 

1.739 

0.036* 

2.981 

0.001* 

-0.82 % 

-2.97 % 

2.968 

0.004* 

3.603  <0.001 

-1.08 % 

-2.04 % 

1.426 

0.055 

2.129 

0.006* 

0.02 % 

-1.9 % 

2.686 

0.007* 

5.150  <0.001 

-0.47 % 

-2.66 % 

2.006 

0.023* 

5.050  <0.001 

Tables A.7- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 175 

 
 
 
 
 
 
 
 
 
Table A.8 Results of Post Hoc Tract Specific Comparisons of Diffusion Kurtosis Imaging Radial 
Diffusivity 

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-
value 
P11 

t-
score 
P2 

q-value 
P21 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

8.422e-04 ± 
3.098e-05 

8.312e-04 ± 
2.900e-05 

8.952e-04 ± 
4.606e-05 

9.084e-04 ± 
5.261e-05 

9.338e-04 ± 
3.814e-05 

8.816e-04 ± 
5.782e-05 

8.787e-04 ± 
3.511e-05 

9.005e-04 ± 
4.544e-05 

8.905e-04 ± 
3.930e-05 

8.753e-04 ± 
3.782e-05 

8.480e-04 ± 
4.596e-05 

8.694e-04 ± 
7.903e-05 

8.698e-04 ± 
3.778e-05 

8.276e-04 ± 
3.046e-05 

8.160e-04 ± 
3.399e-05 

8.687e-04 ± 
3.196e-05 

8.258e-04 ± 
3.205e-05 

8.750e-04 ± 
3.036e-05 

8.673e-04 ± 
3.241e-05 

1.910e-03 ± 
3.849e-04 

1.879e-03 ± 
3.690e-04 

8.072e-04 ± 
8.363e-05 

7.954e-04 ± 
6.566e-05 

8.488e-04 ± 
3.565e-05 

8.417e-04 ± 
4.028e-05 

8.612e-04 ± 
5.113e-05 

8.526e-04 ± 
5.201e-05 

8.348e-04 ± 
3.313e-05 

8.269e-04 ± 
2.550e-05 

8.882e-04 ± 
5.615e-05 

8.970e-04 ± 
5.055e-05 

9.291e-04 ± 
5.964e-05 

8.747e-04 ± 
6.486e-05 

8.786e-04 ± 
4.244e-05 

9.058e-04 ± 
6.995e-05 

8.983e-04 ± 
5.244e-05 

8.883e-04 ± 
5.664e-05 

8.520e-04 ± 
3.726e-05 

8.693e-04 ± 
5.339e-05 

8.719e-04 ± 
3.652e-05 

8.245e-04 ± 
3.115e-05 

8.119e-04 ± 
2.911e-05 

8.661e-04 ± 
4.453e-05 

8.229e-04 ± 
3.218e-05 

8.736e-04 ± 
4.320e-05 

8.661e-04 ± 
3.546e-05 

1.883e-03 ± 
3.814e-04 

1.862e-03 ± 
3.889e-04 

7.985e-04 ± 
5.271e-05 

7.892e-04 ± 
4.163e-05 

8.466e-04 ± 
3.460e-05 

8.375e-04 ± 
3.205e-05 

8.522e-04 ± 
4.061e-05 

8.451e-04 ± 
3.895e-05 

8.099e-04 ± 
4.067e-05 

8.177e-04 ± 
3.143e-05 

8.629e-04 ± 
6.084e-05 

8.917e-04 ± 
7.865e-05 

8.705e-04 ± 
5.066e-05 

8.714e-04 ± 
9.258e-05 

8.584e-04 ± 
5.187e-05 

8.769e-04 ± 
6.721e-05 

8.790e-04 ± 
6.563e-05 

8.696e-04 ± 
7.291e-05 

8.428e-04 ± 
4.536e-05 

8.642e-04 ± 
4.553e-05 

8.597e-04 ± 
4.716e-05 

8.105e-04 ± 
5.637e-05 

8.015e-04 ± 
5.317e-05 

8.385e-04 ± 
6.011e-05 

8.031e-04 ± 
4.801e-05 

8.500e-04 ± 
8.280e-05 

8.499e-04 ± 
7.568e-05 

1.903e-03 ± 
4.074e-04 

1.882e-03 ± 
3.939e-04 

8.028e-04 ± 
7.506e-05 

8.009e-04 ± 
7.779e-05 

8.250e-04 ± 
2.747e-05 

8.273e-04 ± 
3.026e-05 

8.321e-04 ± 
3.420e-05 

8.363e-04 ± 
4.308e-05 

8.002e-04 ± 
4.466e-05 

8.092e-04 ± 
4.166e-05 

8.408e-04 ± 
6.470e-05 

8.724e-04 ± 
7.740e-05 

8.542e-04 ± 
6.646e-05 

8.414e-04 ± 
6.471e-05 

8.427e-04 ± 
6.318e-05 

8.668e-04 ± 
9.904e-05 

8.868e-04 ± 
1.308e-04 

8.757e-04 ± 
1.424e-04 

8.500e-04 ± 
8.756e-05 

8.773e-04 ± 
9.609e-05 

8.577e-04 ± 
8.279e-05 

8.148e-04 ± 
9.633e-05 

7.973e-04 ± 
7.958e-05 

8.342e-04 ± 
9.824e-05 

8.154e-04 ± 
1.009e-04 

8.310e-04 ± 
9.193e-05 

8.298e-04 ± 
8.381e-05 

1.894e-03 ± 
4.438e-04 

1.808e-03 ± 
4.573e-04 

7.909e-04 ± 
9.539e-05 

7.661e-04 ± 
9.179e-05 

8.222e-04 ± 
5.424e-05 

8.221e-04 ± 
4.221e-05 

8.265e-04 ± 
6.365e-05 

8.295e-04 ± 
6.147e-05 

-0.88 % 

-1.19 % 

3.364 

0.013*  3.423 

0.008* 

-0.52 % 

-1.04 % 

1.578 

0.172 

2.120 

0.057 

-0.78 % 

-2.56 % 

2.257 

0.082 

2.897 

0.017* 

-1.26 % 

-2.16 % 

0.985 

0.355 

1.524 

0.122 

-0.5 % 

-1.88 % 

5.342  <0.001  4.354 

0.001* 

-0.78 % 

-3.44 % 

0.530 

0.495 

1.821 

0.085 

0 % 

-1.83 % 

1.792 

0.146 

2.686 

0.022* 

0.59 % 

-1.15 % 

1.617 

0.172 

1.803 

0.085 

0.88 % 

0.89 % 

0.869 

0.366 

0.541 

0.323 

1.49 % 

0.69 % 

0.419 

0.531 

0.548 

0.323 

0.48 % 

0.85 % 

0.388 

0.531 

0.137 

0.407 

-0.01 % 

1.52 % 

0.244 

0.555 

-0.443 

0.334 

0.24 % 

-0.23 % 

0.877 

0.366 

1.012 

0.228 

-0.38 % 

0.52 % 

1.593 

0.172 

0.666 

0.315 

-0.5 % 

-0.52 % 

1.300 

0.249 

1.167 

0.188 

-0.3 % 

-0.52 % 

2.663 

0.052 

1.897 

0.075 

-0.34 % 

1.54 % 

2.190 

0.082 

0.493 

0.327 

-0.15 % 

-2.23 % 

1.884 

0.125 

2.664 

0.022* 

-0.14 % 

-2.37 % 

1.352 

0.244 

2.602 

0.023* 

-1.44 % 

-0.51 % 

0.060 

0.589 

-0.099 

0.407 

-0.94 % 

-3.94 % 

-0.029 

0.589 

0.468 

0.331 

-1.07 % 

-1.48 % 

0.186 

0.572 

0.423 

0.336 

-0.78 % 

-4.34 % 

-0.277 

0.555 

1.469 

0.127 

-0.25 % 

-0.35 % 

2.402 

0.082 

2.203 

0.049* 

-0.49 % 

-0.62 % 

1.293 

0.249 

1.589 

0.112 

-1.05 % 

-0.68 % 

2.080 

0.091 

1.969 

0.074 

-0.88 % 

-0.82 % 

1.114 

0.31 

1.246 

0.174 

 176 

 
 
 
 
 
 
Table A.8 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-
value 
P11 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

7.709e-04 ± 
6.400e-05 

8.491e-04 ± 
3.718e-05 

8.383e-04 ± 
4.026e-05 

8.771e-04 ± 
5.618e-05 

8.644e-04 ± 
6.271e-05 

8.766e-04 ± 
3.370e-05 

8.538e-04 ± 
3.807e-05 

8.344e-04 ± 
3.701e-05 

8.222e-04 ± 
2.993e-05 

8.436e-04 ± 
3.142e-05 

8.262e-04 ± 
2.952e-05 

8.623e-04 ± 
3.992e-05 

8.517e-04 ± 
3.760e-05 

9.040e-04 ± 
5.172e-05 

8.819e-04 ± 
4.788e-05 

8.215e-04 ± 
2.976e-05 

7.966e-04 ± 
2.950e-05 

8.314e-04 ± 
4.660e-05 

8.349e-04 ± 
4.480e-05 

8.532e-04 ± 
4.461e-05 

8.356e-04 ± 
4.836e-05 

8.693e-04 ± 
3.837e-05 

8.433e-04 ± 
3.958e-05 

8.801e-04 ± 
3.850e-05 

8.407e-04 ± 
3.839e-05 

7.697e-04 ± 
4.772e-05 

8.474e-04 ± 
3.487e-05 

8.390e-04 ± 
3.575e-05 

8.731e-04 ± 
5.497e-05 

8.611e-04 ± 
5.522e-05 

8.809e-04 ± 
3.966e-05 

8.567e-04 ± 
3.692e-05 

8.281e-04 ± 
4.502e-05 

8.166e-04 ± 
3.131e-05 

8.413e-04 ± 
3.709e-05 

8.266e-04 ± 
2.818e-05 

8.568e-04 ± 
4.166e-05 

8.470e-04 ± 
3.496e-05 

9.069e-04 ± 
6.034e-05 

8.793e-04 ± 
5.270e-05 

8.158e-04 ± 
4.965e-05 

7.905e-04 ± 
3.606e-05 

8.305e-04 ± 
5.370e-05 

8.297e-04 ± 
4.323e-05 

8.493e-04 ± 
4.268e-05 

8.321e-04 ± 
4.202e-05 

8.707e-04 ± 
3.993e-05 

8.438e-04 ± 
3.587e-05 

8.776e-04 ± 
5.074e-05 

8.388e-04 ± 
4.098e-05 

7.622e-04 ± 
6.892e-05 

8.396e-04 ± 
4.156e-05 

8.361e-04 ± 
4.832e-05 

8.593e-04 ± 
4.323e-05 

8.667e-04 ± 
5.107e-05 

8.763e-04 ± 
6.043e-05 

8.475e-04 ± 
5.643e-05 

8.059e-04 ± 
6.118e-05 

8.140e-04 ± 
7.787e-05 

8.230e-04 ± 
6.739e-05 

8.238e-04 ± 
4.219e-05 

8.461e-04 ± 
5.588e-05 

8.409e-04 ± 
5.194e-05 

9.201e-04 ± 
1.167e-04 

8.841e-04 ± 
8.675e-05 

7.739e-04 ± 
5.671e-05 

7.778e-04 ± 
6.669e-05 

7.872e-04 ± 
4.490e-05 

8.093e-04 ± 
5.486e-05 

8.269e-04 ± 
2.504e-05 

8.240e-04 ± 
2.973e-05 

8.646e-04 ± 
5.316e-05 

8.355e-04 ± 
4.359e-05 

8.617e-04 ± 
7.188e-05 

8.364e-04 ± 
6.896e-05 

7.848e-04 ± 
9.616e-05 

8.338e-04 ± 
5.953e-05 

8.307e-04 ± 
6.177e-05 

8.620e-04 ± 
9.727e-05 

8.627e-04 ± 
7.361e-05 

8.723e-04 ± 
1.045e-04 

8.504e-04 ± 
9.830e-05 

7.971e-04 ± 
8.812e-05 

7.927e-04 ± 
5.724e-05 

8.117e-04 ± 
7.424e-05 

8.184e-04 ± 
6.068e-05 

8.326e-04 ± 
7.016e-05 

8.283e-04 ± 
5.991e-05 

9.083e-04 ± 
1.389e-04 

8.738e-04 ± 
1.198e-04 

7.817e-04 ± 
1.379e-04 

7.659e-04 ± 
8.547e-05 

7.849e-04 ± 
7.971e-05 

7.900e-04 ± 
4.325e-05 

8.290e-04 ± 
6.754e-05 

8.235e-04 ± 
4.740e-05 

8.610e-04 ± 
9.075e-05 

8.366e-04 ± 
7.779e-05 

8.503e-04 ± 
1.108e-04 

8.402e-04 ± 
1.130e-04 

 177 

q-
value 
P21 

0.264 

t-
score 
P2 

-
0.882 

-0.16 % 

2.97 % 

0.461 

0.518 

-0.19 % 

-0.69 % 

0.862 

0.366  1.175 

0.188 

0.09 % 

-0.65 % 

0.178 

0.572  0.697 

0.309 

-0.45 % 

0.31 % 

1.136 

0.31 

0.601 

0.323 

-0.38 % 

-0.45 % 

-0.130 

0.58 

0.407 

-
0.095 

0.5 % 

-0.45 % 

0.026 

0.589  0.518 

0.324 

0.34 % 

0.34 % 

0.518 

0.495  0.404 

0.338 

-0.76 % 

-1.09 % 

2.296 

0.082  1.943 

0.074 

-0.68 % 

-2.61 % 

0.649 

0.44 

2.221  0.049* 

-0.27 % 

-1.37 % 

1.735 

0.157  2.225  0.049* 

0.04 % 

-0.65 % 

0.261 

0.555  0.778 

0.297 

-0.64 % 

-1.59 % 

1.290 

0.249  1.752 

0.091 

-0.56 % 

-1.5 % 

0.916 

0.366  1.602 

0.112 

0.33 % 

-1.29 % 

-0.803 

0.383 

-
0.059 

0.409 

-0.3 % 

-1.17 % 

-0.131 

0.58 

0.272 

0.374 

-0.69 % 

1 % 

4.474  <0.00

1.581 

0.112 

1 

-0.77 % 

-1.53 % 

1.636 

0.172  1.731 

0.092 

-0.11 % 

-0.29 % 

3.296 

0.013*  2.695  0.022* 

-0.62 % 

-2.38 % 

1.883 

0.125  3.256 

0.01* 

-0.46 % 

0.26 % 

2.184 

0.082  1.471 

0.127 

-0.42 % 

-0.07 % 

0.880 

0.366  0.708 

0.309 

0.16 % 

-0.41 % 

0.390 

0.531  0.631 

0.322 

0.06 % 

0.14 % 

0.670 

0.437  0.534 

0.323 

-0.29 % 

-1.32 % 

1.354 

0.244  1.434 

0.132 

-0.22 % 

0.45 % 

0.317 

0.555 

0.407 

-
0.081 

 
 
 
 
 
Table A.8 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-
value 
P11 

t-
score 
P2 

q-
value 
P21 

ST_PREC_left 

ST_PREC_right 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

8.719e-04 ± 
3.638e-05 

8.407e-04 ± 
3.774e-05 

8.546e-04 ± 
3.296e-05 

8.532e-04 ± 
3.367e-05 

8.681e-04 ± 
3.775e-05 

8.483e-04 ± 
3.205e-05 

8.854e-04 ± 
5.808e-05 

8.678e-04 ± 
6.321e-05 

8.839e-04 ± 
4.589e-05 

8.672e-04 ± 
5.007e-05 

8.748e-04 ± 
4.064e-05 

8.381e-04  
4.468e-05 

8.735e-04 ± 
3.971e-05 

8.384e-04 ± 
4.199e-05 

8.774e-04 ± 
3.761e-05 

8.873e-04 ± 
4.435e-05 

8.901e-04 ± 
4.570e-05 

8.706e-04 ± 
4.872e-05 

8.323e-04 ± 
3.522e-05 

8.588e-04 ± 
3.190e-05 

8.652e-04 ± 
5.058e-05 

8.328e-04 ± 
3.523e-05 

8.492e-04 ± 
4.133e-05 

8.467e-04 ± 
3.379e-05 

8.606e-04 ± 
5.403e-05 

8.393e-04 ± 
3.708e-05 

8.808e-04 ± 
5.635e-05 

8.642e-04 ± 
5.531e-05 

8.860e-04 ± 
4.847e-05 

8.667e-04 ± 
4.562e-05 

8.759e-04 ± 
5.389e-05 

8.402e-04 ± 
4.898e-05 

8.671e-04 ± 
5.643e-05 

8.300e-04 ± 
3.974e-05 

8.712e-04 ± 
4.984e-05 

8.780e-04 ± 
4.565e-05 

8.816e-04 ± 
6.655e-05 

8.586e-04 ± 
5.193e-05 

8.361e-04 ± 
3.717e-05 

8.573e-04 ± 
3.760e-05 

8.324e-04 ± 
6.067e-05 

8.217e-04 ± 
5.214e-05 

8.202e-04 ± 
5.619e-05 

8.282e-04 ± 
5.459e-05 

8.250e-04 ± 
5.963e-05 

8.203e-04 ± 
6.524e-05 

8.676e-04 ± 
4.365e-05 

8.694e-04 ± 
5.158e-05 

8.881e-04 ± 
6.674e-05 

8.681e-04 ± 
6.461e-05 

8.789e-04 ± 
9.192e-05 

8.543e-04 ± 
1.009e-04 

8.432e-04 ± 
7.421e-05 

8.268e-04 ± 
6.507e-05 

8.484e-04 ± 
7.055e-05 

8.709e-04 ± 
7.717e-05 

8.578e-04 ± 
7.861e-05 

8.573e-04 ± 
1.056e-04 

8.083e-04 ± 
4.128e-05 

8.412e-04 ± 
5.533e-05 

8.307e-04 ± 
9.639e-05 

8.208e-04 ± 
8.085e-05 

8.096e-04 ± 
6.461e-05 

8.094e-04 ± 
4.894e-05 

7.990e-04 ± 
5.342e-05 

8.066e-04 ± 
6.358e-05 

8.699e-04 ± 
9.549e-05 

8.661e-04 ± 
7.456e-05 

8.837e-04 ± 
1.144e-04 

8.663e-04 ± 
1.052e-04 

8.638e-04 ± 
1.311e-04 

8.571e-04 ± 
1.489e-04 

8.399e-04 ± 
1.161e-04 

8.240e-04 ± 
9.576e-05 

8.334e-04 ± 
8.054e-05 

8.488e-04 ± 
7.828e-05 

8.281e-04 ± 
8.883e-05 

8.376e-04 ± 
1.012e-04 

8.006e-04 ± 
5.181e-05 

8.217e-04 ± 
5.704e-05 

-0.77 % 

-0.2 % 

3.220 

0.014*  1.947 

0.074 

-0.94 % 

-0.11 % 

1.597 

0.172  0.880 

0.264 

-0.62 % 

-1.28 % 

3.071 

0.019*  2.984  0.015* 

-0.76 % 

-2.27 % 

2.226 

0.082  3.531  0.007* 

-0.86 % 

-3.15 % 

3.467 

0.013*  4.057  0.002* 

-1.05 % 

-1.67 % 

2.369 

0.082  2.648  0.022* 

-0.52 % 

0.27 % 

1.107 

0.31 

0.582 

0.323 

-0.42 % 

-0.38 % 

-0.091 

0.589 

-
0.113 

0.407 

0.25 % 

-0.49 % 

-0.289 

0.555  0.122 

0.407 

-0.06 % 

-0.21 % 

-0.061 

0.589  0.022 

0.415 

0.12 % 

-1.72 % 

-0.255 

0.555  0.562 

0.323 

0.25 % 

0.32 % 

-0.931 

0.366 

0.304 

-
0.743 

-0.73 % 

-0.38 % 

2.161 

0.082  1.325 

0.156 

-1 % 

-0.34 % 

0.851 

0.366  0.384 

0.339 

-0.71 % 

-1.77 % 

2.178 

0.082  2.333  0.043* 

-1.04 % 

-2.54 % 

1.078 

0.316  1.924 

0.074 

-0.95 % 

-3.46 % 

2.070 

0.091  2.644  0.022* 

-1.38 % 

-2.3 % 

0.717 

0.421  1.144 

0.19 

0.46 % 

-0.95 % 

2.265 

0.082  3.096  0.014* 

-0.17 % 

-2.32 % 

1.618 

0.172  2.991  0.015* 

Tables A.8- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 178 

 
 
 
 
 
 
 
 
 
Table A.9 Results of Post Hoc Tract Specific Comparisons of Neurite Orientation Dispersion and 
Density Imaging Orientation Dispersion Index 

TractSeg 
Abbv. 

AF_left 

AF_right 

ATR_left 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI 

Controls 

0.351 ± 0.015  0.354 ± 0.012  0.362 ± 0.016  0.365 ± 0.017 

0.9 % 

0.346 ± 0.010  0.350 ± 0.011  0.356 ± 0.014  0.362 ± 0.020 

1.19 % 

0.320 ± 0.013  0.322 ± 0.012  0.327 ± 0.012  0.331 ± 0.015 

0.58 % 

ATR_right 

0.323 ± 0.011  0.326 ± 0.012  0.332 ± 0.012  0.337 ± 0.018 

0.82 % 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

0.283 ± 0.018  0.286 ± 0.022  0.286 ± 0.016  0.301 ± 0.042 

1.28 % 

0.247 ± 0.017  0.252 ± 0.035  0.248 ± 0.022  0.263 ± 0.046 

1.96 % 

0.303 ± 0.012  0.305 ± 0.012  0.313 ± 0.016  0.319 ± 0.018 

0.76 % 

0.309 ± 0.017  0.312 ± 0.018  0.319 ± 0.026  0.327 ± 0.026 

1.04 % 

0.299 ± 0.015  0.302 ± 0.017  0.310 ± 0.019  0.317 ± 0.023 

1.16 % 

0.285 ± 0.015  0.290 ± 0.016  0.299 ± 0.020  0.305 ± 0.025 

1.63 % 

0.322 ± 0.015  0.324 ± 0.015  0.333 ± 0.019  0.337 ± 0.021 

0.62 % 

0.319 ± 0.013  0.322 ± 0.016  0.325 ± 0.015  0.333 ± 0.033 

0.65 % 

0.312 ± 0.012  0.315 ± 0.012  0.322 ± 0.015  0.328 ± 0.019 

0.73 % 

CG_left 

CG_right 

CST_left 

0.327 ± 0.018  0.326 ± 0.020  0.337 ± 0.022  0.344 ± 0.026 

-0.28 % 

0.325 ± 0.021  0.326 ± 0.017  0.337 ± 0.019  0.345 ± 0.024 

0.36 % 

0.265 ± 0.012  0.266 ± 0.012  0.277 ± 0.014  0.281 ± 0.015 

0.52 % 

CST_right 

0.270 ± 0.011  0.272 ± 0.012  0.279 ± 0.013  0.287 ± 0.020 

0.82 % 

FPT_left 

0.267 ± 0.012  0.268 ± 0.012  0.281 ± 0.017  0.287 ± 0.018 

0.32 % 

FPT_right 

0.265 ± 0.011  0.267 ± 0.012  0.279 ± 0.013  0.283 ± 0.016 

0.62 % 

0.86 % 

1.71 % 

1.29 % 

1.47 % 

5.3 % 

5.9 % 

2.09 % 

2.49 % 

2.33 % 

1.84 % 

1.22 % 

2.34 % 

2.03 % 

2.16 % 

2.42 % 

1.67 % 

2.65 % 

1.94 % 

1.43 % 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

-2.501  <0.001 

-2.973 

<0.001 

-3.137  <0.001 

-3.167 

<0.001 

-1.939  <0.001 

-2.740 

<0.001 

-2.696  <0.001 

-2.840 

<0.001 

-0.616 

0.004* 

-1.904 

<0.001 

-0.192 

0.006* 

-1.027 

<0.001 

-2.653  <0.001 

-3.697 

<0.001 

-1.809  <0.001 

-2.626 

<0.001 

-2.363  <0.001 

-2.865 

<0.001 

-2.985  <0.001 

-2.866 

<0.001 

-2.284  <0.001 

-2.740 

<0.001 

-1.561 

0.001* 

-2.001 

<0.001 

-2.506  <0.001 

-3.467 

<0.001 

-1.812  <0.001 

-3.050 

<0.001 

-2.019  <0.001 

-3.646 

<0.001 

-3.393  <0.001 

-4.399 

<0.001 

-3.044  <0.001 

-3.755 

<0.001 

-3.771  <0.001 

-5.007 

<0.001 

-3.976  <0.001 

-4.269 

<0.001 

FX_left 

FX_right 

ICP_left 

0.140 ± 0.043  0.147 ± 0.048  0.154 ± 0.062  0.178 ± 0.079 

5.17 % 

15.52 % 

-1.050 

0.002* 

-1.968 

<0.001 

0.135 ± 0.041  0.142 ± 0.043  0.152 ± 0.058  0.167 ± 0.079 

5.24 % 

10.19 % 

-1.299 

0.002* 

-1.692 

<0.001 

0.357 ± 0.018  0.359 ± 0.018  0.376 ± 0.023  0.374 ± 0.026 

0.82 % 

-0.62 % 

-3.507  <0.001 

-2.560 

<0.001 

ICP_right 

0.372 ± 0.017  0.376 ± 0.018  0.390 ± 0.018  0.385 ± 0.023 

1.28 % 

-1.51 % 

-3.804  <0.001 

-1.560 

<0.001 

IFO_left 

0.319 ± 0.011  0.321 ± 0.013  0.326 ± 0.013  0.335 ± 0.028 

0.67 % 

IFO_right 

0.323 ± 0.009  0.326 ± 0.012  0.329 ± 0.014  0.334 ± 0.025 

0.85 % 

ILF_left 

0.320 ± 0.012  0.324 ± 0.021  0.325 ± 0.011  0.339 ± 0.037 

1.36 % 

ILF_right 

0.324 ± 0.015  0.328 ± 0.022  0.326 ± 0.018  0.336 ± 0.032 

1.07 % 

MCP 

0.344 ± 0.015  0.347 ± 0.015  0.359 ± 0.020  0.359 ± 0.023 

0.87 % 

MLF_left 

0.348 ± 0.015  0.349 ± 0.014  0.357 ± 0.017  0.359 ± 0.017 

0.25 % 

MLF_right 

0.344 ± 0.015  0.347 ± 0.015  0.354 ± 0.018  0.359 ± 0.020 

0.91 % 

OR_left 

0.309 ± 0.013  0.311 ± 0.014  0.316 ± 0.012  0.326 ± 0.032 

0.56 % 

OR_right 

0.305 ± 0.012  0.306 ± 0.013  0.309 ± 0.018  0.316 ± 0.029 

0.38 % 

POPT_left 

0.294 ± 0.014  0.296 ± 0.013  0.306 ± 0.015  0.308 ± 0.015 

0.5 % 

POPT_right 

0.289 ± 0.013  0.291 ± 0.014  0.297 ± 0.015  0.301 ± 0.016 

0.67 % 

SCP_left 

0.300 ± 0.013  0.303 ± 0.014  0.319 ± 0.016  0.319 ± 0.021 

0.86 % 

2.96 % 

1.63 % 

4.52 % 

3.19 % 

0.22 % 

0.58 % 

1.42 % 

3.12 % 

2.38 % 

0.73 % 

1.33 % 

0.06 % 

-2.116  <0.001 

-2.950 

<0.001 

-1.874  <0.001 

-1.894 

<0.001 

-1.394 

0.001* 

-2.133 

<0.001 

-0.328 

0.005* 

-1.221 

<0.001 

-3.277  <0.001 

-2.711 

<0.001 

-2.060  <0.001 

-2.587 

<0.001 

-2.245  <0.001 

-2.558 

<0.001 

-1.926  <0.001 

-2.760 

<0.001 

-1.058 

0.002* 

-2.050 

<0.001 

-2.877  <0.001 

-3.201 

<0.001 

-1.915  <0.001 

-2.306 

<0.001 

-4.760  <0.001 

-3.689 

<0.001 

SCP_right 

0.303 ± 0.010  0.307 ± 0.014  0.320 ± 0.016  0.319 ± 0.020 

1.3 % 

-0.5 % 

-5.145  <0.001 

-2.801 

<0.001 

SLF_III_left 

0.340 ± 0.017  0.344 ± 0.015  0.353 ± 0.024  0.356 ± 0.018 

1.33 % 

SLF_III_right  0.337 ± 0.012  0.343 ± 0.017  0.352 ± 0.019  0.360 ± 0.029 

1.87 % 

0.91 % 

2.05 % 

-2.429  <0.001 

-2.626 

<0.001 

-4.080  <0.001 

-3.021 

<0.001 

SLF_II_left 

0.363 ± 0.016  0.362 ± 0.016  0.372 ± 0.017  0.371 ± 0.027 

-0.26 % 

-0.3 % 

-1.939  <0.001 

-1.728 

<0.001 

SLF_II_right 

0.364 ± 0.015  0.365 ± 0.015  0.376 ± 0.017  0.379 ± 0.017 

0.35 % 

0.7 % 

-2.805  <0.001 

-3.091 

<0.001 

SLF_I_left 

0.360 ± 0.017  0.360 ± 0.017  0.371 ± 0.016  0.377 ± 0.017 

0 % 

SLF_I_right 

0.352 ± 0.016  0.353 ± 0.018  0.362 ± 0.016  0.364 ± 0.018 

0.12 % 

STR_left 

0.270 ± 0.012  0.272 ± 0.012  0.283 ± 0.013  0.292 ± 0.025 

0.76 % 

STR_right 

0.272 ± 0.014  0.275 ± 0.013  0.283 ± 0.014  0.287 ± 0.016 

1.04 % 

ST_FO_left 

0.295 ± 0.017  0.299 ± 0.025  0.301 ± 0.018  0.309 ± 0.035 

1.52 % 

ST_FO_right 

0.287 ± 0.016  0.290 ± 0.026  0.294 ± 0.017  0.302 ± 0.028 

1.19 % 

ST_OCC_left  0.312 ± 0.013  0.313 ± 0.015  0.321 ± 0.014  0.330 ± 0.035 

0.47 % 

1.51 % 

0.66 % 

3.44 % 

1.62 % 

2.47 % 

2.81 % 

2.78 % 

-2.184  <0.001 

-3.391 

<0.001 

-2.028  <0.001 

-2.284 

<0.001 

-3.328  <0.001 

-4.455 

<0.001 

-2.602  <0.001 

-3.180 

<0.001 

-1.336 

0.001* 

-1.252 

<0.001 

-1.538 

0.001* 

-1.637 

<0.001 

-2.445  <0.001 

-2.895 

<0.001 

 179 

 
 
 
 
 
 
Table A.9 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

ST_OCC_right 

0.315 ± 0.013  0.317 ± 0.014  0.321 ± 0.020  0.326 ± 0.026 

0.65 % 

1.83 % 

-1.251 

0.002* 

-1.828  <0.001 

ST_PAR_left 

0.327 ± 0.014  0.328 ± 0.014  0.337 ± 0.015  0.341 ± 0.015 

0.21 % 

0.98 % 

-2.450  <0.001 

-3.133  <0.001 

ST_PAR_right 

0.319 ± 0.013  0.322 ± 0.013  0.327 ± 0.014  0.331 ± 0.016 

0.86 % 

1.18 % 

-2.230  <0.001 

-2.437  <0.001 

ST_POSTC_left 

0.305 ± 0.015  0.308 ± 0.014  0.319 ± 0.012  0.324 ± 0.013 

0.87 % 

1.51 % 

-3.206  <0.001 

-3.989  <0.001 

ST_POSTC_right  0.301 ± 0.014  0.305 ± 0.014  0.310 ± 0.013  0.315 ± 0.018 

1.29 % 

1.65 % 

-2.212  <0.001 

-2.360  <0.001 

ST_PREC_left 

0.306 ± 0.013  0.309 ± 0.011  0.318 ± 0.011  0.326 ± 0.013 

0.78 % 

2.53 % 

-3.227  <0.001 

-5.235  <0.001 

ST_PREC_right 

0.305 ± 0.010  0.308 ± 0.012  0.314 ± 0.012  0.320 ± 0.015 

0.98 % 

1.94 % 

-2.909  <0.001 

-3.433  <0.001 

ST_PREF_left 

0.312 ± 0.012  0.313 ± 0.011  0.322 ± 0.015  0.330 ± 0.017 

0.41 % 

2.42 % 

-2.927  <0.001 

-4.780  <0.001 

ST_PREF_right 

0.312 ± 0.010  0.314 ± 0.012  0.323 ± 0.012  0.328 ± 0.016 

0.76 % 

1.69 % 

-3.558  <0.001 

-3.951  <0.001 

ST_PREM_left 

0.325 ± 0.015  0.327 ± 0.014  0.339 ± 0.016  0.348 ± 0.016 

0.57 % 

2.53 % 

-3.181  <0.001 

-5.007  <0.001 

ST_PREM_right 

0.318 ± 0.014  0.320 ± 0.014  0.330 ± 0.014  0.338 ± 0.023 

0.64 % 

2.42 % 

-2.980  <0.001 

-3.827  <0.001 

T_OCC_left 

0.313 ± 0.012  0.314 ± 0.014  0.319 ± 0.011  0.329 ± 0.031 

0.56 % 

3.07 % 

-1.764  <0.001 

-2.679  <0.001 

T_OCC_right 

0.311 ± 0.012  0.312 ± 0.013  0.314 ± 0.017  0.322 ± 0.028 

0.34 % 

2.31 % 

-0.956 

0.003* 

-2.021  <0.001 

T_PAR_left 

0.326 ± 0.013  0.326 ± 0.013  0.335 ± 0.013  0.338 ± 0.016 

0.22 % 

1.02 % 

-2.363  <0.001 

-3.001  <0.001 

T_PAR_right 

0.316 ± 0.013  0.317 ± 0.014  0.323 ± 0.012  0.327 ± 0.014 

0.51 % 

1.19 % 

-1.973  <0.001 

-2.448  <0.001 

T_POSTC_left 

0.304 ± 0.014  0.307 ± 0.014  0.319 ± 0.015  0.322 ± 0.016 

0.95 % 

0.84 % 

-3.600  <0.001 

-3.514  <0.001 

T_POSTC_right 

0.301 ± 0.015  0.304 ± 0.016  0.311 ± 0.018  0.316 ± 0.020 

1.13 % 

1.64 % 

-2.308  <0.001 

-2.496  <0.001 

T_PREC_left 

0.303 ± 0.012  0.305 ± 0.012  0.314 ± 0.012  0.321 ± 0.015 

0.65 % 

2.32 % 

-3.090  <0.001 

-4.501  <0.001 

T_PREC_right 

0.308 ± 0.011  0.310 ± 0.012  0.316 ± 0.011  0.322 ± 0.014 

0.6 % 

2.01 % 

-2.424  <0.001 

-3.406  <0.001 

T_PREF_left 

0.307 ± 0.011  0.308 ± 0.010  0.318 ± 0.014  0.324 ± 0.017 

0.32 % 

2.13 % 

-2.936  <0.001 

-4.834  <0.001 

T_PREF_right 

0.314 ± 0.010  0.315 ± 0.011  0.324 ± 0.012  0.329 ± 0.014 

0.56 % 

1.56 % 

-3.341  <0.001 

-4.082  <0.001 

T_PREM_left 

0.312 ± 0.013  0.314 ± 0.012  0.324 ± 0.012  0.330 ± 0.015 

0.54 % 

1.82 % 

-3.139  <0.001 

-4.301  <0.001 

T_PREM_right 

0.309 ± 0.013  0.310 ± 0.013  0.318 ± 0.012  0.325 ± 0.018 

0.28 % 

2.21 % 

-2.423  <0.001 

-3.829  <0.001 

UF_left 

0.300 ± 0.014  0.303 ± 0.022  0.307 ± 0.017  0.319 ± 0.035 

1.2 % 

3.62 % 

-1.783  <0.001 

-2.126  <0.001 

0.297 ± 0.014  0.301 ± 0.021  0.305 ± 0.015  0.315 ± 0.032 

UF_right 
Tables A.9- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

-2.081  <0.001 

1.49 % 

3.28 % 

-2.064  <0.001 

 180 

 
 
 
 
 
 
 
 
 
Table A.10 Results of Post Hoc Tract Specific Comparisons of Neurite Orientation Dispersion 
and Density Imaging Neurite Density Index 

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

AF_left 

AF_right 

ATR_left 

0.482 ± 0.022  0.481 ± 0.021  0.493 ± 0.030 

0.497 ± 0.025 

-0.25 % 

0.75 % 

-1.531 

0.038* 

-2.550  <0.001 

0.503 ± 0.023  0.501 ± 0.019  0.507 ± 0.032 

0.508 ± 0.034 

-0.37 % 

0.12 % 

-0.547 

0.087 

-0.995 

0.003* 

0.478 ± 0.019  0.479 ± 0.020  0.499 ± 0.029 

0.503 ± 0.024 

0.13 % 

0.73 % 

-3.394 

0.005* 

-4.034  <0.001 

ATR_right 

0.480 ± 0.019  0.480 ± 0.018  0.495 ± 0.023 

0.500 ± 0.025 

0.04 % 

1.03 % 

-2.638 

0.009* 

-3.553  <0.001 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

SCP_left 

SCP_right 

0.401 ± 0.026  0.396 ± 0.026  0.426 ± 0.028 

0.445 ± 0.024 

-1.18 % 

4.44 % 

-3.218 

0.005* 

-6.782  <0.001 

0.485 ± 0.031  0.476 ± 0.037  0.506 ± 0.031 

0.496 ± 0.050 

-1.73 % 

-1.84 % 

-2.316 

0.015* 

-1.790  <0.001 

0.490 ± 0.023  0.488 ± 0.023  0.502 ± 0.033 

0.503 ± 0.029 

-0.36 % 

0.09 % 

-1.669 

0.033* 

-2.101  <0.001 

0.493 ± 0.023  0.494 ± 0.024  0.509 ± 0.045 

0.511 ± 0.034 

0.1 % 

0.36 % 

-1.882 

0.024* 

-2.287  <0.001 

0.543 ± 0.026  0.544 ± 0.029  0.548 ± 0.041 

0.554 ± 0.038 

0.03 % 

0.538 ± 0.024  0.537 ± 0.025  0.546 ± 0.036 

0.547 ± 0.035 

-0.3 % 

1.05 % 

0.27 % 

-0.566 

-0.980 

0.497 ± 0.021  0.494 ± 0.020  0.498 ± 0.027 

0.497 ± 0.027 

-0.62 % 

-0.16 % 

-0.146 

0.514 ± 0.023  0.512 ± 0.021  0.516 ± 0.023 

0.514 ± 0.033 

-0.35 % 

-0.27 % 

-0.210 

0.501 ± 0.020  0.499 ± 0.020  0.508 ± 0.030 

0.508 ± 0.028 

-0.38 % 

0.08 % 

-1.020 

0.087 

0.061 

0.118 

0.113 

0.061 

-1.198 

0.002* 

-1.395 

0.002* 

-0.539 

0.004* 

-0.271 

0.006* 

-1.472 

0.001* 

0.473 ± 0.025  0.473 ± 0.025  0.484 ± 0.033 

0.482 ± 0.035 

-0.1 % 

-0.58 % 

-1.425 

0.042* 

-1.151 

0.002* 

0.482 ± 0.025  0.480 ± 0.022  0.488 ± 0.033 

0.493 ± 0.030 

-0.46 % 

1.13 % 

-0.678 

0.079 

-1.924  <0.001 

0.598 ± 0.021  0.597 ± 0.023  0.607 ± 0.034 

0.619 ± 0.034 

-0.03 % 

1.96 % 

-1.341 

0.047* 

-2.912  <0.001 

0.600 ± 0.022  0.599 ± 0.023  0.607 ± 0.033 

0.610 ± 0.042 

-0.18 % 

0.48 % 

-1.004 

0.061 

-1.405 

0.002* 

0.559 ± 0.022  0.559 ± 0.024  0.576 ± 0.038 

0.580 ± 0.042 

0.06 % 

0.62 % 

-2.301 

0.015* 

-2.552  <0.001 

0.559 ± 0.020  0.558 ± 0.022  0.572 ± 0.034 

0.579 ± 0.033 

-0.21 % 

1.12 % 

-1.900 

0.024* 

-2.978  <0.001 

0.451 ± 0.053  0.451 ± 0.048  0.463 ± 0.039 

0.481 ± 0.052 

-0.13 % 

3.88 % 

-0.818 

0.071 

-2.227  <0.001 

0.447 ± 0.053  0.454 ± 0.056  0.467 ± 0.042 

0.471 ± 0.044 

1.52 % 

0.69 % 

-1.360 

0.046* 

-1.102 

0.002* 

0.591 ± 0.028  0.589 ± 0.036  0.588 ± 0.042 

0.594 ± 0.060 

-0.36 % 

0.577 ± 0.021  0.573 ± 0.026  0.571 ± 0.046 

0.584 ± 0.048 

-0.64 % 

0.92 % 

2.26 % 

0.342 

0.775 

0.106 

0.072 

-0.375 

0.005* 

-1.171 

0.002* 

0.482 ± 0.022  0.477 ± 0.020  0.491 ± 0.022 

0.491 ± 0.032 

-1.04 % 

-0.01 % 

-1.453 

0.041* 

-2.187  <0.001 

0.486 ± 0.022  0.483 ± 0.020  0.493 ± 0.021 

0.497 ± 0.035 

-0.69 % 

0.76 % 

-1.019 

0.061 

-2.016  <0.001 

0.467 ± 0.022  0.463 ± 0.021  0.478 ± 0.023 

0.483 ± 0.029 

-0.81 % 

1.02 % 

-1.656 

0.033* 

-3.025  <0.001 

0.471 ± 0.021  0.469 ± 0.020  0.480 ± 0.026 

0.486 ± 0.033 

-0.51 % 

1.25 % 

-1.307 

0.047* 

-2.600  <0.001 

0.617 ± 0.023  0.615 ± 0.027  0.610 ± 0.046 

0.622 ± 0.046 

-0.44 % 

0.470 ± 0.021  0.467 ± 0.020  0.474 ± 0.024 

0.477 ± 0.023 

-0.56 % 

0.477 ± 0.020  0.474 ± 0.018  0.477 ± 0.028 

0.479 ± 0.030 

-0.65 % 

1.91 % 

0.79 % 

0.31 % 

0.891 

-0.623 

-0.015 

0.065 

0.084 

0.128 

-0.761 

0.003* 

-1.804  <0.001 

-0.786 

0.003* 

0.487 ± 0.022  0.483 ± 0.022  0.497 ± 0.020 

0.501 ± 0.033 

-0.77 % 

0.86 % 

-1.539 

0.038* 

-2.530  <0.001 

0.493 ± 0.022  0.490 ± 0.022  0.499 ± 0.026 

0.505 ± 0.033 

-0.58 % 

1.2 % 

POPT_left 

0.531 ± 0.019  0.528 ± 0.019  0.537 ± 0.026 

0.541 ± 0.028 

-0.67 % 

0.88 % 

-0.903 

-0.928 

0.065 

0.064 

-2.144  <0.001 

-2.274  <0.001 

POPT_right 

0.538 ± 0.017  0.537 ± 0.019  0.545 ± 0.025 

0.548 ± 0.032 

-0.27 % 

0.55 % 

-1.261 

0.049* 

-1.803  <0.001 

0.575 ± 0.021  0.575 ± 0.028  0.589 ± 0.039 

0.601 ± 0.046 

-0.01 % 

2.01 % 

-1.837 

0.025* 

-2.770  <0.001 

0.572 ± 0.016  0.570 ± 0.022  0.577 ± 0.041 

0.591 ± 0.039 

-0.26 % 

SLF_III_left 

0.513 ± 0.026  0.508 ± 0.026  0.518 ± 0.040 

0.518 ± 0.034 

-1.03 % 

2.32 % 

0.04 % 

-0.801 

-0.569 

SLF_III_right 

0.517 ± 0.025  0.513 ± 0.024  0.519 ± 0.036 

0.514 ± 0.048 

-0.81 % 

-0.99 % 

-0.213 

SLF_II_left 

0.488 ± 0.023  0.490 ± 0.026  0.490 ± 0.039 

0.498 ± 0.033 

0.31 % 

SLF_II_right 

0.488 ± 0.023  0.488 ± 0.024  0.488 ± 0.038 

0.495 ± 0.026 

0.02 % 

SLF_I_left 

0.464 ± 0.022  0.465 ± 0.027  0.466 ± 0.036 

0.472 ± 0.029 

0.22 % 

SLF_I_right 

0.481 ± 0.023  0.483 ± 0.026  0.485 ± 0.038 

0.492 ± 0.028 

0.39 % 

1.57 % 

1.28 % 

1.25 % 

1.46 % 

-0.243 

-0.117 

-0.268 

-0.535 

0.071 

0.087 

0.113 

0.113 

-2.767  <0.001 

-1.321 

0.002* 

-0.081 

0.007* 

-1.049 

0.002* 

0.12 

-1.039 

0.002* 

0.112 

0.087 

-0.869 

0.003* 

-1.258 

0.002* 

STR_left 

STR_right 

0.590 ± 0.025  0.589 ± 0.025  0.605 ± 0.028 

0.609 ± 0.049 

-0.06 % 

0.57 % 

-2.137 

0.018* 

-2.207  <0.001 

0.598 ± 0.021  0.593 ± 0.023  0.607 ± 0.025 

0.613 ± 0.030 

-0.74 % 

0.98 % 

-1.509 

0.038* 

-2.933  <0.001 

ST_FO_left 

0.465 ± 0.026  0.462 ± 0.026  0.492 ± 0.029 

0.496 ± 0.030 

-0.7 % 

0.93 % 

-3.389 

0.005* 

-4.450  <0.001 

ST_FO_right 

0.467 ± 0.027  0.464 ± 0.029  0.486 ± 0.023 

0.486 ± 0.046 

-0.64 % 

-0.02 % 

-2.437 

0.013* 

-2.292  <0.001 

ST_OCC_left 

0.485 ± 0.022  0.481 ± 0.021  0.493 ± 0.020 

0.497 ± 0.033 

-0.89 % 

0.78 % 

-1.197 

0.052 

-2.324  <0.001 

 181 

 
 
 
 
 
Table A.10 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

ST_OCC_right 

0.492 ± 0.022  0.489 ± 0.021  0.497 ± 0.023 

0.503 ± 0.032 

-0.65 % 

ST_PAR_left 

0.494 ± 0.020  0.491 ± 0.019  0.501 ± 0.023 

0.504 ± 0.023 

-0.61 % 

ST_PAR_right 

0.509 ± 0.019  0.508 ± 0.018  0.515 ± 0.024 

0.517 ± 0.029 

-0.33 % 

1.24 % 

0.67 % 

0.28 % 

-0.720 

-1.112 

-1.034 

0.076 

0.059 

0.061 

-2.106  <0.001 

-2.281  <0.001 

-1.548 

0.001* 

ST_POSTC_left 

0.529 ± 0.019  0.529 ± 0.021  0.547 ± 0.023 

0.553 ± 0.023 

-0.14 % 

1.04 % 

-3.006 

0.007* 

-3.994  <0.001 

ST_POSTC_right  0.546 ± 0.021  0.547 ± 0.019  0.560 ± 0.025 

0.562 ± 0.033 

0.18 % 

0.3 % 

-2.220 

0.017* 

-2.323  <0.001 

ST_PREC_left 

0.547 ± 0.021  0.547 ± 0.023  0.563 ± 0.028 

0.571 ± 0.029 

0.04 % 

1.52 % 

-2.417 

0.013* 

-3.553  <0.001 

ST_PREC_right 

0.558 ± 0.024  0.558 ± 0.022  0.571 ± 0.029 

0.573 ± 0.033 

-0.04 % 

0.41 % 

-1.725 

0.031* 

-2.153  <0.001 

ST_PREF_left 

0.503 ± 0.022  0.502 ± 0.021  0.521 ± 0.031 

0.522 ± 0.031 

-0.1 % 

0.09 % 

-2.713 

0.009* 

-2.931  <0.001 

ST_PREF_right 

0.504 ± 0.023  0.503 ± 0.020  0.519 ± 0.026 

0.524 ± 0.028 

-0.28 % 

0.91 % 

-2.184 

0.017* 

-3.349  <0.001 

ST_PREM_left 

0.502 ± 0.021  0.503 ± 0.021  0.521 ± 0.033 

0.527 ± 0.027 

0.25 % 

1.2 % 

-2.783 

0.008* 

-3.871  <0.001 

ST_PREM_right 

0.522 ± 0.023  0.522 ± 0.021  0.542 ± 0.034 

0.540 ± 0.033 

0.03 % 

-0.3 % 

-2.627 

0.009* 

-2.592  <0.001 

T_OCC_left 

0.486 ± 0.022  0.482 ± 0.022  0.496 ± 0.020 

0.499 ± 0.032 

-0.71 % 

0.7 % 

-1.605 

0.035* 

-2.463  <0.001 

T_OCC_right 

0.492 ± 0.021  0.489 ± 0.021  0.498 ± 0.025 

0.504 ± 0.033 

-0.52 % 

T_PAR_left 

0.497 ± 0.019  0.494 ± 0.020  0.503 ± 0.025 

0.506 ± 0.025 

-0.5 % 

T_PAR_right 

0.510 ± 0.018  0.508 ± 0.019  0.516 ± 0.023 

0.517 ± 0.027 

-0.35 % 

1.11 % 

0.63 % 

0.35 % 

-0.983 

-1.024 

-1.015 

0.061 

0.061 

0.061 

-2.143  <0.001 

-2.011  <0.001 

-1.579 

0.001* 

T_POSTC_left 

0.529 ± 0.020  0.528 ± 0.022  0.541 ± 0.027 

0.548 ± 0.026 

-0.24 % 

1.23 % 

-1.920 

0.024* 

-3.101  <0.001 

T_POSTC_right 

0.539 ± 0.020  0.539 ± 0.021  0.546 ± 0.027 

0.549 ± 0.034 

0 % 

T_PREC_left 

0.555 ± 0.021  0.556 ± 0.024  0.569 ± 0.031 

0.579 ± 0.031 

0.2 % 

0.47 % 

1.85 % 

-1.199 

-2.061 

0.052 

0.02* 

-1.483 

0.001* 

-3.210  <0.001 

T_PREC_right 

0.559 ± 0.022  0.559 ± 0.023  0.568 ± 0.030 

0.571 ± 0.033 

-0.03 % 

0.61 % 

-1.322 

0.047* 

-1.770  <0.001 

T_PREF_left 

0.508 ± 0.021  0.509 ± 0.022  0.527 ± 0.032 

0.528 ± 0.032 

0.17 % 

0.21 % 

-2.774 

0.008* 

-2.802  <0.001 

T_PREF_right 

0.503 ± 0.020  0.503 ± 0.020  0.517 ± 0.027 

0.523 ± 0.025 

-0.1 % 

1.27 % 

-2.182 

0.017* 

-3.522  <0.001 

T_PREM_left 

0.507 ± 0.020  0.510 ± 0.022  0.529 ± 0.034 

0.537 ± 0.027 

0.53 % 

1.61 % 

-3.164 

0.005* 

-4.262  <0.001 

T_PREM_right 

0.521 ± 0.019  0.521 ± 0.022  0.539 ± 0.033 

0.541 ± 0.028 

0.07 % 

0.37 % 

-2.768 

0.008* 

-3.003  <0.001 

UF_left 

0.438 ± 0.026  0.432 ± 0.026  0.452 ± 0.025 

0.458 ± 0.019 

-1.24 % 

1.28 % 

-1.952 

0.023* 

-3.719  <0.001 

0.430 ± 0.024  0.426 ± 0.023  0.443 ± 0.018 

UF_right 
Tables A.10- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

0.451 ± 0.026 

-0.96 % 

1.74 % 

0.023* 

-1.983 

-3.723  <0.001 

 182 

 
 
 
 
 
 
 
 
 
 
TractSeg 
Abbv. 

AF_left 

AF_right 

ATR_left 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

1.903 

2.203 

1.160 

1.056 

0.649 

2.731 

2.077 

1.698 

1.390 

0.540 

-0.318 

1.526 

1.241 

0.697 

0.081 

0.055 

0.218 

0.238 

0.318 

0.026* 

0.064 

0.105 

0.158 

0.318 

0.353 

0.13 

0.198 

0.318 

Table A.11 Results of Post Hoc Tract Specific Comparisons of Neurite Orientation Dispersion 
and Density Imaging Free Water Fraction 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

0.071 ± 0.012 

0.066 ± 0.015 

0.058 ± 0.009 

0.055 ± 0.014 

-7.44 % 

-5.23 % 

3.838  <0.001 

2.569 

0.031* 

0.076 ± 0.013 

0.072 ± 0.013 

0.069 ± 0.013 

0.065 ± 0.012 

-4.97 % 

-6.23 % 

0.090 ± 0.016 

0.086 ± 0.022 

0.082 ± 0.018 

0.072 ± 0.018 

-4.3 % 

-11.69 % 

ATR_right 

0.097 ± 0.019 

0.091 ± 0.019 

0.092 ± 0.024 

0.084 ± 0.025 

-5.92 % 

-8.32 % 

0.069 ± 0.026 

0.061 ± 0.025 

0.054 ± 0.021 

0.053 ± 0.029 

-10.8 % 

-0.49 % 

1.716 

1.623 

0.826 

2.059 

0.039* 

0.039* 

0.089 

0.024* 

0.100 ± 0.026 

0.092 ± 0.031 

0.106 ± 0.032 

0.087 ± 0.023 

-7.51 % 

-18.55 % 

-0.865 

0.089 

0.099 ± 0.013 

0.097 ± 0.020 

0.092 ± 0.015 

0.082 ± 0.018 

-2.58 % 

-10.93 % 

0.114 ± 0.021 

0.113 ± 0.031 

0.103 ± 0.026 

0.095 ± 0.032 

-0.46 % 

-8.03 % 

0.141 ± 0.019 

0.142 ± 0.028 

0.130 ± 0.023 

0.127 ± 0.041 

0.37 % 

-2.04 % 

0.131 ± 0.019 

0.133 ± 0.026 

0.126 ± 0.027 

0.121 ± 0.044 

1.86 % 

-3.46 % 

0.081 ± 0.017 

0.080 ± 0.015 

0.078 ± 0.012 

0.077 ± 0.025 

-0.65 % 

-0.53 % 

0.094 ± 0.027 

0.093 ± 0.019 

0.093 ± 0.019 

0.095 ± 0.029 

-1.28 % 

2.33 % 

0.097 ± 0.014 

0.096 ± 0.016 

0.093 ± 0.013 

0.088 ± 0.023 

-1.29 % 

-5.23 % 

0.060 ± 0.012 

0.059 ± 0.014 

0.057 ± 0.014 

0.053 ± 0.021 

-2.33 % 

-6.26 % 

0.060 ± 0.011 

0.056 ± 0.012 

0.054 ± 0.009 

0.053 ± 0.025 

-6.13 % 

-2.81 % 

1.948 

1.596 

2.061 

0.835 

0.597 

0.198 

1.035 

0.901 

1.673 

0.028* 

0.039* 

0.024* 

0.089 

0.104 

0.118 

0.078 

0.089 

0.039* 

0.168 ± 0.017 

0.164 ± 0.025 

0.143 ± 0.016 

0.139 ± 0.027 

-2.3 % 

-2.3 % 

5.086  <0.001 

3.413 

0.013* 

CST_right 

0.145 ± 0.015 

0.142 ± 0.017 

0.131 ± 0.016 

0.130 ± 0.025 

-2.41 % 

-0.92 % 

FPT_left 

0.144 ± 0.015 

0.141 ± 0.022 

0.128 ± 0.019 

0.118 ± 0.021 

-1.77 % 

-7.88 % 

FPT_right 

0.140 ± 0.013 

0.136 ± 0.017 

0.128 ± 0.019 

0.120 ± 0.022 

-2.35 % 

-6.47 % 

FX_left 

FX_right 

ICP_left 

0.501 ± 0.108 

0.490 ± 0.103 

0.485 ± 0.103 

0.476 ± 0.123 

-2.24 % 

-2 % 

0.491 ± 0.104 

0.483 ± 0.108 

0.482 ± 0.104 

0.451 ± 0.136 

-1.76 % 

-6.31 % 

0.095 ± 0.029 

0.089 ± 0.016 

0.089 ± 0.013 

0.092 ± 0.018 

-5.86 % 

2.98 % 

ICP_right 

0.082 ± 0.026 

0.077 ± 0.017 

0.081 ± 0.019 

0.081 ± 0.025 

-5.93 % 

-0.41 % 

IFO_left 

0.074 ± 0.014 

0.069 ± 0.015 

0.068 ± 0.012 

0.065 ± 0.014 

-6.55 % 

-4.94 % 

IFO_right 

0.072 ± 0.016 

0.067 ± 0.014 

0.069 ± 0.014 

0.067 ± 0.020 

-5.81 % 

-2.47 % 

ILF_left 

0.070 ± 0.020 

0.063 ± 0.018 

0.063 ± 0.015 

0.061 ± 0.019 

-10.73 % 

-2.8 % 

ILF_right 

0.067 ± 0.023 

0.061 ± 0.019 

0.064 ± 0.018 

0.065 ± 0.026 

-8.04 % 

2.2 % 

MCP 

0.095 ± 0.027 

0.092 ± 0.019 

0.085 ± 0.012 

0.097 ± 0.026 

-2.89 % 

14.42 % 

MLF_left 

0.064 ± 0.015 

0.061 ± 0.014 

0.061 ± 0.012 

0.060 ± 0.020 

-4.84 % 

-2.67 % 

MLF_right 

0.062 ± 0.017 

0.060 ± 0.015 

0.061 ± 0.013 

0.059 ± 0.019 

-2.78 % 

-2.82 % 

OR_left 

OR_right 

0.090 ± 0.024 

0.085 ± 0.022 

0.086 ± 0.016 

0.087 ± 0.028 

-5.8 % 

0.89 % 

0.086 ± 0.026 

0.083 ± 0.022 

0.089 ± 0.022 

0.088 ± 0.030 

-3.77 % 

-0.46 % 

-0.395 

POPT_left 

0.121 ± 0.017 

0.119 ± 0.019 

0.117 ± 0.020 

0.113 ± 0.033 

-1.27 % 

-3.2 % 

POPT_right 

0.113 ± 0.018 

0.113 ± 0.018 

0.111 ± 0.020 

0.109 ± 0.029 

-0.26 % 

-1.29 % 

SCP_left 

0.114 ± 0.016 

0.109 ± 0.015 

0.100 ± 0.009 

0.100 ± 0.019 

-4.04 % 

0.06 % 

SCP_right 

0.104 ± 0.013 

0.099 ± 0.014 

0.097 ± 0.016 

0.097 ± 0.017 

-4.74 % 

-0.8 % 

SLF_III_left 

0.092 ± 0.015 

0.086 ± 0.020 

0.078 ± 0.012 

0.070 ± 0.021 

-6.36 % 

-10.49 % 

SLF_III_right 

0.083 ± 0.015 

0.080 ± 0.015 

0.080 ± 0.014 

0.072 ± 0.015 

-3.91 % 

-9.31 % 

SLF_II_left 

0.079 ± 0.015 

0.077 ± 0.020 

0.068 ± 0.011 

0.066 ± 0.022 

-2.78 % 

-4.03 % 

SLF_II_right 

0.072 ± 0.014 

0.070 ± 0.015 

0.065 ± 0.012 

0.062 ± 0.017 

-2.79 % 

-5.41 % 

SLF_I_left 

0.083 ± 0.022 

0.085 ± 0.025 

0.087 ± 0.035 

0.083 ± 0.047 

1.82 % 

-5.55 % 

SLF_I_right 

0.083 ± 0.020 

0.084 ± 0.021 

0.086 ± 0.027 

0.082 ± 0.039 

0.49 % 

-4.34 % 

0.721 

0.410 

3.251 

1.708 

3.368 

0.861 

2.561 

1.708 

-0.574 

-0.371 

 183 

3.285 

3.472 

2.664 

0.503 

0.313 

0.713 

0.183 

1.536 

0.578 

1.323 

0.437 

1.402 

0.648 

0.315 

0.637 

0.002* 

0.002* 

0.008* 

0.109 

0.113 

0.099 

0.118 

0.042* 

0.104 

0.058 

0.111 

0.053 

0.104 

0.113 

0.104 

0.111 

0.099 

0.111 

0.002* 

0.039* 

0.002* 

0.089 

0.009* 

0.039* 

0.104 

0.111 

2.230 

3.852 

3.196 

0.464 

0.966 

-0.551 

-0.609 

1.059 

0.025 

0.286 

-0.683 

-0.871 

0.286 

0.333 

-0.287 

-0.849 

0.922 

0.573 

2.017 

0.627 

2.913 

1.846 

1.961 

1.863 

0.251 

0.255 

0.055 

0.007* 

0.013* 

0.328 

0.265 

0.318 

0.318 

0.238 

0.395 

0.353 

0.318 

0.276 

0.353 

0.353 

0.353 

0.276 

0.275 

0.318 

0.069 

0.318 

0.02* 

0.083 

0.074 

0.083 

0.353 

0.353 

 
 
 
 
 
 
Table A.11 (cont’d)  

TractSeg Abbv. 

STR_left 

STR_right 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

t-
score 
P2 

q-value 
P21 

0.141 ± 0.016 

0.136 ± 0.026  0.117 ± 0.021  0.113 ± 0.030 

-3.47 % 

-4.08 % 

4.787  <0.001 

3.158 

0.013* 

2.042 

0.927 

0.131 ± 0.017 

0.123 ± 0.020  0.119 ± 0.025  0.111 ± 0.027 

-5.49 % 

-7.14 % 

ST_FO_left 

0.061 ± 0.016 

0.057 ± 0.023  0.056 ± 0.017  0.053 ± 0.022 

-5.49 % 

-5.37 % 

ST_FO_right 

0.066 ± 0.017 

0.060 ± 0.019  0.066 ± 0.021  0.056 ± 0.018 

-7.66 % 

-14.49 % 

-0.103 

ST_OCC_left 

0.080 ± 0.018 

0.074 ± 0.017  0.071 ± 0.012  0.072 ± 0.020 

-6.34 % 

1.42 % 

ST_OCC_right 

0.073 ± 0.020 

0.069 ± 0.018  0.070 ± 0.014  0.071 ± 0.023 

-5.05 % 

1.85 % 

ST_PAR_left 

0.091 ± 0.017 

0.089 ± 0.018  0.090 ± 0.020  0.086 ± 0.030 

-2.12 % 

-4.22 % 

ST_PAR_right 

0.087 ± 0.018 

0.086 ± 0.016  0.086 ± 0.016  0.083 ± 0.026 

-1.01 % 

-3.16 % 

ST_POSTC_left 

0.124 ± 0.019 

0.121 ± 0.025  0.117 ± 0.030  0.109 ± 0.038 

-2.35 % 

-6.97 % 

1.620 

0.440 

0.285 

0.200 

1.124 

ST_POSTC_right  0.111 ± 0.019 

0.111 ± 0.021  0.115 ± 0.029  0.110 ± 0.040 

-0.23 % 

-3.84 % 

-0.543 

ST_PREC_left 

0.133 ± 0.019 

0.128 ± 0.026  0.113 ± 0.019  0.109 ± 0.031 

-3.7 % 

-3.51 % 

ST_PREC_right 

0.122 ± 0.018 

0.117 ± 0.018  0.115 ± 0.022  0.108 ± 0.027 

-3.55 % 

-5.86 % 

ST_PREF_left 

0.094 ± 0.013 

0.091 ± 0.020  0.083 ± 0.015  0.074 ± 0.016 

-4.1 % 

-10.22 % 

ST_PREF_right 

0.095 ± 0.013 

0.090 ± 0.016  0.088 ± 0.018  0.079 ± 0.017 

-4.96 % 

-10.48 % 

ST_PREM_left 

0.099 ± 0.018 

0.094 ± 0.027  0.083 ± 0.017  0.071 ± 0.017 

-4.37 % 

-14.08 % 

ST_PREM_right 

0.104 ± 0.016 

0.099 ± 0.019  0.096 ± 0.019  0.084 ± 0.023 

-4.73 % 

-12.23 % 

T_OCC_left 

0.092 ± 0.024 

0.086 ± 0.022  0.088 ± 0.016  0.088 ± 0.028 

-5.71 % 

0.41 % 

3.629 

1.166 

2.935 

1.681 

3.119 

1.594 

0.589 

0.024* 

2.072 

0.089 

0.123 

0.644 

0.742 

0.039* 

0.424 

0.111 

-0.473 

0.114 

0.118 

0.072 

0.106 

0.557 

0.545 

1.541 

0.111 

0.064 

0.318 

0.311 

0.337 

0.328 

0.318 

0.318 

0.13 

0.378 

0.001* 

2.478 

0.032* 

0.069 

1.519 

0.13 

0.004* 

2.925 

0.02* 

0.039* 

2.489 

0.032* 

0.003* 

3.287 

0.013* 

0.039* 

2.555 

0.031* 

0.104 

-0.263 

T_OCC_right 

0.085 ± 0.026 

0.082 ± 0.022  0.088 ± 0.022  0.087 ± 0.030 

-3.74 % 

-0.36 % 

-0.382 

0.111 

-0.848 

T_PAR_left 

0.098 ± 0.020 

0.096 ± 0.021  0.100 ± 0.024  0.095 ± 0.037 

-1.45 % 

-4.59 % 

-0.312 

T_PAR_right 

0.097 ± 0.021 

0.095 ± 0.019  0.098 ± 0.025  0.095 ± 0.034 

-1.29 % 

T_POSTC_left 

0.120 ± 0.019 

0.119 ± 0.025  0.120 ± 0.038  0.111 ± 0.047 

-1.31 % 

-3.7 % 

-7.8 % 

-0.261 

-0.026 

0.113 

0.115 

0.13 

0.183 

0.118 

0.877 

T_POSTC_right 

0.105 ± 0.019 

0.106 ± 0.022  0.114 ± 0.039  0.110 ± 0.049 

0.56 % 

-3.25 % 

-1.220 

0.067 

-0.510 

T_PREC_left 

0.139 ± 0.019 

0.135 ± 0.028  0.120 ± 0.023  0.116 ± 0.037 

-3.17 % 

-3.68 % 

T_PREC_right 

0.120 ± 0.018 

0.115 ± 0.019  0.114 ± 0.025  0.108 ± 0.031 

-4.04 % 

-5.42 % 

T_PREF_left 

0.107 ± 0.014 

0.103 ± 0.022  0.096 ± 0.019  0.086 ± 0.021 

-3.38 % 

-10.47 % 

T_PREF_right 

0.106 ± 0.016 

0.101 ± 0.018  0.100 ± 0.023  0.091 ± 0.024 

-5.11 % 

-9.32 % 

T_PREM_left 

0.110 ± 0.019 

0.106 ± 0.029  0.098 ± 0.022  0.086 ± 0.027 

-3.74 % 

-12.54 % 

T_PREM_right 

0.111 ± 0.018 

0.105 ± 0.021  0.106 ± 0.029  0.094 ± 0.031 

-5.76 % 

-10.69 % 

UF_left 

0.043 ± 0.015 

0.040 ± 0.016  0.041 ± 0.012  0.039 ± 0.019 

-6.96 % 

-5.22 % 

3.244 

1.041 

2.473 

1.186 

2.170 

0.879 

0.448 

0.002* 

2.186 

0.078 

1.146 

0.011* 

2.804 

0.023* 

0.069 

1.808 

0.087 

0.021* 

2.567 

0.031* 

0.089 

0.111 

1.514 

0.218 

0.13 

0.359 

0.051 ± 0.014 

0.046 ± 0.014  0.050 ± 0.019  0.046 ± 0.022 

UF_right 
Tables A.11- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

-8.95 % 

-8.54 % 

0.049 

0.109 

0.123 

0.393 

 184 

0.353 

0.276 

0.365 

0.378 

0.276 

0.323 

0.055 

0.218 

 
 
 
 
 
 
 
 
 
Table A.12 Results of Post Hoc Tract Specific Comparisons of Fixel-Based Analysis Fiber Density 

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI   Controls  

mTBI 

Controls 

Percent Change 

t-
score 
P1 

q-
value 
P11 

t-
score 
P2 

q-value 
P2 

AF_left 

AF_right 

ATR_left 

0.285 ± 0.018 

0.286 ± 0.014 

0.296 ± 0.014 

0.296 ± 0.011 

0.39 % 

-0.12 % 

-2.371  0.005* 

-2.641  <0.001 

0.294 ± 0.015 

0.293 ± 0.014 

0.297 ± 0.015 

0.297 ± 0.013 

-0.23 % 

0.28 % 

-0.713 

0.07 

-1.145  0.003* 

0.298 ± 0.014 

0.298 ± 0.016 

0.316 ± 0.016 

0.318 ± 0.014 

0 % 

0.57 % 

-4.689  <0.001 

-4.699  <0.001 

ATR_right 

0.291 ± 0.014 

0.292 ± 0.015 

0.306 ± 0.014 

0.310 ± 0.014 

0.18 % 

1.4 % 

-3.629  <0.001 

-4.445  <0.001 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

0.269 ± 0.025 

0.271 ± 0.024 

0.295 ± 0.027 

0.302 ± 0.025 

0.81 % 

2.26 % 

-3.661  <0.001 

-4.481  <0.001 

0.378 ± 0.024 

0.374 ± 0.026 

0.392 ± 0.024 

0.393 ± 0.023 

-1.04 % 

0.02 % 

-2.138  0.008* 

-2.578  <0.001 

0.331 ± 0.018 

0.330 ± 0.019 

0.344 ± 0.018 

0.346 ± 0.015 

-0.21 % 

0.55 % 

-2.654  0.003* 

-3.143  <0.001 

0.326 ± 0.022 

0.328 ± 0.024 

0.344 ± 0.029 

0.346 ± 0.028 

0.6 % 

0.55 % 

-2.776  0.002* 

-2.601  <0.001 

0.366 ± 0.023 

0.367 ± 0.023 

0.386 ± 0.020 

0.389 ± 0.019 

0.14 % 

0.6 % 

-3.235  <0.001 

-3.513  <0.001 

0.370 ± 0.021 

0.370 ± 0.021 

0.385 ± 0.025 

0.389 ± 0.024 

-0.11 % 

0.8 % 

-2.548  0.004* 

-3.060  <0.001 

0.352 ± 0.023 

0.352 ± 0.019 

0.359 ± 0.015 

0.357 ± 0.016 

-0.15 % 

-0.55 % 

-1.043 

0.05* 

-0.967  0.003* 

0.330 ± 0.029 

0.331 ± 0.020 

0.333 ± 0.014 

0.332 ± 0.016 

0.4 % 

-0.17 % 

-0.466  0.088 

-0.303  0.006* 

0.343 ± 0.018 

0.343 ± 0.016 

0.353 ± 0.016 

0.354 ± 0.016 

-0.1 % 

0.2 % 

-2.029 

0.01* 

-2.418  <0.001 

0.314 ± 0.016 

0.313 ± 0.017 

0.324 ± 0.016 

0.326 ± 0.013 

-0.24 % 

0.58 % 

-2.162  0.008* 

-2.672  <0.001 

0.314 ± 0.017 

0.313 ± 0.018 

0.319 ± 0.019 

0.322 ± 0.014 

-0.33 % 

0.95 % 

-0.937  0.055 

-1.808  <0.001 

0.428 ± 0.016 

0.431 ± 0.016 

0.449 ± 0.017 

0.454 ± 0.017 

0.67 % 

1.04 % 

-4.459  <0.001 

-5.065  <0.001 

0.437 ± 0.016 

0.437 ± 0.014 

0.451 ± 0.015 

0.455 ± 0.018 

0.18 % 

0.95 % 

-3.180  0.001* 

-4.240  <0.001 

0.398 ± 0.015 

0.400 ± 0.015 

0.416 ± 0.015 

0.420 ± 0.016 

0.51 % 

1.06 % 

-4.142  <0.001 

-4.532  <0.001 

0.400 ± 0.014 

0.401 ± 0.013 

0.413 ± 0.016 

0.418 ± 0.016 

0.24 % 

1.16 % 

-3.315  <0.001 

-4.487  <0.001 

0.210 ± 0.052 

0.213 ± 0.057 

0.237 ± 0.051 

0.241 ± 0.056 

1.45 % 

1.87 % 

-1.845  0.014* 

-1.778  <0.001 

0.221 ± 0.054 

0.224 ± 0.059 

0.236 ± 0.058 

0.241 ± 0.055 

0.302 ± 0.017 

0.302 ± 0.012 

0.302 ± 0.014 

0.305 ± 0.015 

1.2 % 

0.1 % 

2.37 % 

-0.966  0.054 

-1.095  0.003* 

0.95 % 

-0.072  0.119 

-0.833  0.004* 

0.277 ± 0.016 

0.277 ± 0.015 

0.277 ± 0.010 

0.278 ± 0.017 

-0.2 % 

0.25 % 

0.082 

0.119 

-0.215  0.007* 

0.313 ± 0.017 

0.314 ± 0.015 

0.323 ± 0.012 

0.323 ± 0.013 

0.15 % 

0.05 % 

-2.080  0.009* 

-2.174  <0.001 

0.313 ± 0.016 

0.313 ± 0.014 

0.319 ± 0.011 

0.321 ± 0.012 

0.03 % 

0.57 % 

-1.284  0.035* 

-1.877  <0.001 

0.288 ± 0.022 

0.290 ± 0.015 

0.293 ± 0.013 

0.295 ± 0.013 

0.84 % 

0.37 % 

-1.019 

0.05 

-1.059  0.003* 

0.291 ± 0.019 

0.291 ± 0.016 

0.296 ± 0.012 

0.299 ± 0.013 

0.02 % 

1.01 % 

-0.913  0.055 

-1.659  0.001* 

0.315 ± 0.014 

0.315 ± 0.012 

0.315 ± 0.011 

0.317 ± 0.010 

MLF_left 

0.308 ± 0.022 

0.308 ± 0.016 

0.315 ± 0.016 

0.313 ± 0.018 

0.1 % 

0.3 % 

0.52 % 

-0.067  0.119 

-0.476  0.005* 

-0.57 % 

-1.218  0.039* 

-0.933  0.003* 

MLF_right 

0.325 ± 0.018 

0.324 ± 0.017 

0.328 ± 0.015 

0.325 ± 0.017 

-0.46 % 

-0.75 % 

-0.430  0.088 

-0.241  0.007* 

OR_left 

OR_right 

0.316 ± 0.021 

0.318 ± 0.021 

0.325 ± 0.014 

0.328 ± 0.014 

0.49 % 

0.92 % 

-1.527  0.025* 

-1.839  <0.001 

0.335 ± 0.022 

0.336 ± 0.019 

0.338 ± 0.012 

0.340 ± 0.014 

0.35 % 

0.74 % 

-0.438  0.088 

-0.734  0.004* 

POPT_left 

0.379 ± 0.015 

0.380 ± 0.014 

0.390 ± 0.018 

0.391 ± 0.019 

0.2 % 

0.4 % 

-2.323  0.006* 

-2.605  <0.001 

POPT_right 

0.402 ± 0.017 

0.402 ± 0.016 

0.411 ± 0.015 

0.412 ± 0.018 

0 % 

0.29 % 

-1.867  0.014* 

-2.192  <0.001 

SCP_left 

SCP_right 

0.342 ± 0.012 

0.344 ± 0.012 

0.351 ± 0.013 

0.356 ± 0.012 

0.44 % 

1.66 % 

-2.489  0.004* 

-3.817  <0.001 

0.339 ± 0.011 

0.340 ± 0.012 

0.344 ± 0.011 

0.348 ± 0.013 

0.15 % 

1.2 % 

-1.652 

0.02* 

-2.634  <0.001 

SLF_III_left 

0.304 ± 0.017 

0.305 ± 0.015 

0.313 ± 0.019 

0.314 ± 0.019 

0.1 % 

0.2 % 

-1.802  0.015* 

-2.057  <0.001 

SLF_III_right 

0.300 ± 0.018 

0.299 ± 0.018 

0.298 ± 0.017 

0.299 ± 0.017 

-0.29 % 

0.16 % 

0.451 

0.088 

0.174 

0.007* 

SLF_II_left 

0.284 ± 0.017 

0.284 ± 0.017 

0.288 ± 0.021 

0.289 ± 0.020 

0.1 % 

0.37 % 

-0.914  0.055 

-1.056  0.003* 

SLF_II_right 

0.281 ± 0.017 

0.281 ± 0.018 

0.283 ± 0.013 

0.283 ± 0.015 

-0.19 % 

-0.12 % 

-0.476  0.088 

-0.495  0.005* 

SLF_I_left 

0.276 ± 0.014 

0.275 ± 0.014 

0.278 ± 0.016 

0.278 ± 0.017 

-0.26 % 

-0.15 % 

-0.682  0.072 

-0.731  0.004* 

SLF_I_right 

0.289 ± 0.015 

0.289 ± 0.015 

0.295 ± 0.014 

0.294 ± 0.015 

0 % 

-0.39 % 

-1.479  0.027* 

-1.207  0.002* 

STR_left 

STR_right 

0.416 ± 0.017 

0.418 ± 0.018 

0.432 ± 0.022 

0.439 ± 0.020 

0.64 % 

1.65 % 

-3.115  0.001* 

-3.971  <0.001 

0.434 ± 0.015 

0.433 ± 0.015 

0.444 ± 0.023 

0.449 ± 0.023 

-0.02 % 

1.16 % 

-2.197  0.008* 

-3.287  <0.001 

ST_FO_left 

0.320 ± 0.019 

0.319 ± 0.019 

0.339 ± 0.015 

0.341 ± 0.018 

-0.29 % 

0.64 % 

-3.527  <0.001 

-4.112  <0.001 

ST_FO_right 

0.335 ± 0.021 

0.334 ± 0.019 

0.350 ± 0.016 

0.355 ± 0.016 

-0.52 % 

1.24 % 

-2.617  0.003* 

-4.146  <0.001 

ST_OCC_left 

0.326 ± 0.023 

0.327 ± 0.020 

0.335 ± 0.013 

0.337 ± 0.015 

0.46 % 

0.59 % 

-1.458  0.027* 

-1.716  0.001* 

 185 

 
 
 
 
 
 
Table A.12 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI  

Controls  

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P2 

ST_OCC_right 

0.334 ± 0.022  0.335 ± 0.018  0.339 ± 0.013  0.341 ± 0.015 

0.32 % 

0.45 % 

-0.847 

0.059 

-1.068 

0.003* 

ST_PAR_left 

0.335 ± 0.016  0.335 ± 0.015  0.345 ± 0.016  0.345 ± 0.018 

0.18 % 

0.08 % 

-2.152 

0.008* 

-2.133  <0.001 

ST_PAR_right 

0.355 ± 0.017  0.354 ± 0.016  0.361 ± 0.013  0.361 ± 0.016 

-0.23 % 

0 % 

-1.406 

0.029* 

-1.594 

0.001* 

ST_POSTC_left 

0.338 ± 0.013  0.340 ± 0.014  0.355 ± 0.017  0.358 ± 0.018 

0.54 % 

0.85 % 

-4.379  <0.001 

-4.323  <0.001 

ST_POSTC_right  0.361 ± 0.015  0.360 ± 0.014  0.374 ± 0.016  0.376 ± 0.019 

-0.2 % 

0.53 % 

-2.958 

0.002* 

-3.487  <0.001 

ST_PREC_left 

0.346 ± 0.013  0.348 ± 0.014  0.365 ± 0.015  0.368 ± 0.016 

0.57 % 

0.87 % 

-5.135  <0.001 

-5.065  <0.001 

ST_PREC_right 

0.363 ± 0.014  0.363 ± 0.013  0.376 ± 0.015  0.379 ± 0.015 

0.13 % 

0.75 % 

-3.407  <0.001 

-4.122  <0.001 

ST_PREF_left 

0.329 ± 0.013  0.329 ± 0.015  0.344 ± 0.013  0.345 ± 0.015 

0.04 % 

0.51 % 

-3.916  <0.001 

-3.937  <0.001 

ST_PREF_right 

0.330 ± 0.013  0.329 ± 0.012  0.340 ± 0.013  0.344 ± 0.013 

-0.12 % 

1.13 % 

-2.961 

0.002* 

-4.224  <0.001 

ST_PREM_left 

0.308 ± 0.014  0.309 ± 0.016  0.320 ± 0.024  0.322 ± 0.023 

0.38 % 

0.64 % 

-2.544 

0.004* 

-2.576  <0.001 

ST_PREM_right 

0.322 ± 0.016  0.324 ± 0.016  0.335 ± 0.019  0.337 ± 0.017 

0.53 % 

0.61 % 

-2.825 

0.002* 

-2.917  <0.001 

T_OCC_left 

0.310 ± 0.021  0.312 ± 0.020  0.319 ± 0.013  0.322 ± 0.014 

0.46 % 

0.96 % 

-1.510 

0.026* 

-1.892  <0.001 

T_OCC_right 

0.328 ± 0.021  0.329 ± 0.019  0.331 ± 0.012  0.333 ± 0.014 

0.37 % 

T_PAR_left 

0.335 ± 0.015  0.335 ± 0.015  0.344 ± 0.017  0.345 ± 0.018 

0.06 % 

0.61 % 

0.29 % 

-0.588 

-2.031 

0.079 

0.01* 

-0.824 

0.004* 

-2.227  <0.001 

T_PAR_right 

0.362 ± 0.017  0.361 ± 0.017  0.367 ± 0.015  0.368 ± 0.017 

-0.17 % 

0.24 % 

-1.139 

0.043* 

-1.472 

0.002* 

T_POSTC_left 

0.357 ± 0.014  0.358 ± 0.015  0.371 ± 0.021  0.374 ± 0.021 

0.25 % 

1.04 % 

-3.056 

0.001* 

-3.558  <0.001 

T_POSTC_right 

0.372 ± 0.017  0.371 ± 0.017  0.383 ± 0.020  0.387 ± 0.021 

-0.27 % 

0.86 % 

-2.372 

0.005* 

-3.213  <0.001 

T_PREC_left 

0.365 ± 0.015  0.367 ± 0.015  0.386 ± 0.016  0.390 ± 0.016 

0.52 % 

1 % 

-4.968  <0.001 

-5.187  <0.001 

T_PREC_right 

0.377 ± 0.015  0.377 ± 0.014  0.390 ± 0.016  0.393 ± 0.016 

0.13 % 

0.94 % 

-3.046 

0.001* 

-3.944  <0.001 

T_PREF_left 

0.333 ± 0.013  0.333 ± 0.014  0.349 ± 0.013  0.352 ± 0.014 

0.14 % 

0.69 % 

-4.522  <0.001 

-4.582  <0.001 

T_PREF_right 

0.322 ± 0.012  0.322 ± 0.013  0.334 ± 0.013  0.338 ± 0.013 

0 % 

1.19 % 

-3.503  <0.001 

-4.490  <0.001 

T_PREM_left 

0.324 ± 0.013  0.326 ± 0.016  0.340 ± 0.021  0.343 ± 0.020 

0.5 % 

0.78 % 

-3.815  <0.001 

-3.674  <0.001 

T_PREM_right 

0.335 ± 0.015  0.337 ± 0.017  0.352 ± 0.019  0.356 ± 0.018 

0.63 % 

0.94 % 

-3.820  <0.001 

-3.839  <0.001 

UF_left 

0.299 ± 0.020  0.299 ± 0.017  0.313 ± 0.015  0.310 ± 0.024 

-0.16 % 

-0.77 % 

-2.490 

0.004* 

-2.158  <0.001 

0.295 ± 0.016  0.293 ± 0.017  0.307 ± 0.011  0.307 ± 0.020 

UF_right 
Tables A.12- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

-0.41 % 

0.27 % 

0.002* 

-2.812 

-2.828  <0.001 

 186 

 
 
 
 
 
 
 
 
 
 
Table A.13 Results of Post Hoc Tract Specific Comparisons of Fixel-Based Analysis Log Fiber 
Cross Section 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI 

Controls 

TractSeg 
Abbv. 

t-score 
P1 

q-
value 
P11 

t-score 
P2 

q-value 
P21 

AF_left 

0.035 ± 0.073 

0.035 ± 0.070 

0.056 ± 0.111 

0.068 ± 0.125 

-0.26 % 

21.44 % 

-0.908  0.269 

-1.404 

0.003* 

AF_right 

0.046 ± 0.070 

0.045 ± 0.073 

0.078 ± 0.101 

0.087 ± 0.121 

-1.42 % 

11.55 % 

-1.469  0.216 

-1.732 

0.003* 

ATR_left 

0.003 ± 0.082 

-6.888e-03 ± 
8.403e-02 

0.026 ± 0.102 

0.028 ± 0.121 

-299.02 % 

8.83 % 

-0.920  0.269 

-1.339 

0.004* 

ATR_right 

0.014 ± 0.079 

0.008 ± 0.080 

0.052 ± 0.091 

0.056 ± 0.106 

-43.33 % 

6.57 % 

-1.679  0.216 

-1.973 

0.003* 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

0.005 ± 0.083 

0.010 ± 0.072 

0.056 ± 0.089 

0.073 ± 0.085 

104.23 % 

29.87 % 

-2.162  0.211 

-2.978  <0.001 

0.013 ± 0.089 

0.003 ± 0.092 

0.077 ± 0.107 

0.081 ± 0.101 

-75.6 % 

4.94 % 

-2.441  0.141 

-2.950  <0.001 

0.020 ± 0.069 

0.015 ± 0.071 

0.054 ± 0.093 

0.057 ± 0.102 

-28.35 % 

6.43 % 

-1.593  0.216 

-1.932 

0.003* 

0.023 ± 0.080 

0.026 ± 0.088 

0.028 ± 0.103 

0.039 ± 0.119 

10.96 % 

39.9 % 

-0.193  0.419 

-0.493 

0.006* 

0.003 ± 0.072 

0.013 ± 0.075 

0.047 ± 0.078 

0.064 ± 0.097 

286.67 % 

37.54 % 

-2.110  0.211 

-2.297 

0.002* 

-1.106e-02 ± 
6.943e-02 

8.253e-04 ± 
7.889e-02 

-2.758e-03 ± 
7.171e-02 

0.007 ± 0.103 

0.017 ± 0.113 

-75.08 % 

156.28 % 

-0.808  0.269 

-0.845 

0.004* 

0.006 ± 0.073 

0.006 ± 0.106 

0.014 ± 0.110 

612 % 

122.82 % 

-0.220  0.415 

-0.336 

0.007* 

0.057 ± 0.079 

0.053 ± 0.080 

0.043 ± 0.103 

0.060 ± 0.115 

-6.36 % 

40.95 % 

0.595 

0.316 

-0.276 

0.007* 

0.013 ± 0.061 

0.014 ± 0.063 

0.032 ± 0.091 

0.040 ± 0.099 

8.12 % 

25.81 % 

-0.987  0.269 

-1.292 

0.004* 

CG_left 

0.008 ± 0.063 

0.006 ± 0.063 

0.026 ± 0.101 

0.022 ± 0.092 

-26.31 % 

-14.76 % 

-0.877  0.269 

-0.822 

0.004* 

CG_right 

0.008 ± 0.063 

0.004 ± 0.064 

0.024 ± 0.094 

0.025 ± 0.090 

-47.98 % 

5.64 % 

-0.836  0.269 

-1.088 

0.004* 

CST_left 

CST_right 

-1.905e-02 ± 
7.521e-02 

-3.508e-03 ± 
7.713e-02 

-8.765e-03 ± 
7.238e-02 

0.039 ± 0.085 

0.051 ± 0.095 

-53.99 % 

29.98 % 

-2.684  0.099 

-2.750  <0.001 

0.003 ± 0.076 

0.035 ± 0.093 

0.041 ± 0.096 

-174.57 % 

17.98 % 

-1.699  0.216 

-1.714 

0.003* 

FPT_left 

0.004 ± 0.068 

0.006 ± 0.065 

0.032 ± 0.073 

0.039 ± 0.081 

52.09 % 

23.18 % 

-1.432  0.216 

-1.702 

0.003* 

FPT_right 

0.001 ± 0.064 

FX_left 

FX_right 

-7.329e-02 ± 
1.008e-01 

-7.471e-02 ± 
1.073e-01 

4.733e-04 ± 
6.411e-02 

-7.435e-02 ± 
1.209e-01 

-7.727e-02 ± 
1.254e-01 

0.030 ± 0.091 

0.033 ± 0.090 

-65.91 % 

11.89 % 

-1.435  0.216 

-1.665 

0.003* 

-1.159e-01 ± 
1.119e-01 

-1.160e-01 ± 
1.090e-01 

-1.177e-01 ± 
1.174e-01 

-1.159e-01 ± 
1.192e-01 

1.44 % 

1.52 % 

1.469 

0.216  1.282 

0.004* 

3.42 % 

-0.08 % 

1.363 

0.225  1.106 

0.004* 

ICP_left 

0.025 ± 0.091 

0.024 ± 0.097 

0.054 ± 0.097 

0.049 ± 0.090 

ICP_right 

0.023 ± 0.081 

0.018 ± 0.085 

0.036 ± 0.094 

0.038 ± 0.103 

IFO_left 

0.035 ± 0.074 

0.028 ± 0.074 

0.044 ± 0.105 

0.055 ± 0.120 

-4.04 % 

-24.1 % 

-18.8 % 

-9.26 % 

-1.107  0.269 

-0.916 

0.004* 

7.15 % 

-0.537  0.333 

-0.833 

0.004* 

23 % 

-0.436  0.366 

-1.109 

0.004* 

IFO_right 

0.047 ± 0.067 

0.042 ± 0.068 

0.066 ± 0.099 

0.077 ± 0.111 

-10.62 % 

16.59 % 

-0.882  0.269 

-1.554 

0.003* 

ILF_left 

0.069 ± 0.085 

0.067 ± 0.087 

0.070 ± 0.107 

0.091 ± 0.113 

-2.51 % 

28.83 % 

-0.068  0.447 

-0.904 

0.004* 

ILF_right 

0.091 ± 0.082 

0.090 ± 0.082 

0.091 ± 0.104 

0.107 ± 0.107 

-1 % 

18.01 % 

-0.011  0.456 

-0.712 

0.005* 

 187 

 
 
 
 
 
 
Table A.13 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-
value 
P11 

t-score 
P2 

q-value 
P21 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

0.028 ± 0.094 

0.027 ± 0.097  0.044 ± 0.105  0.045 ± 0.107 

-5.45 % 

0.86 % 

-0.599 

0.316 

-0.652 

0.005* 

0.017 ± 0.082 

0.020 ± 0.075  0.026 ± 0.110  0.034 ± 0.113 

18.51 % 

32.39 % 

-0.350 

0.395 

-0.587 

0.005* 

0.020 ± 0.078 

0.025 ± 0.073  0.026 ± 0.102  0.032 ± 0.108 

28.69 % 

24.39 % 

-0.270 

0.408 

-0.308 

0.007* 

0.037 ± 0.080 

0.038 ± 0.080  0.036 ± 0.098  0.052 ± 0.114 

1.13 % 

44.33 % 

0.051 

0.447 

-0.574 

0.005* 

0.035 ± 0.075 

0.034 ± 0.074  0.042 ± 0.104  0.055 ± 0.118 

-1.84 % 

30.6 % 

-0.324 

0.399 

-0.881 

0.004* 

-3.043e-02 ± 
7.537e-02 

-2.240e-02 ± 
6.974e-02 

-6.398e-03 ± 
9.750e-02 

-4.521e-04 ± 
1.052e-01 

-2.630e-02 ± 
7.626e-02 

-1.965e-02 ± 
7.247e-02 

-4.322e-03 ± 
1.042e-01 

-9.075e-05 ± 
1.045e-01 

-26.39 % 

-92.93 % 

-1.063 

0.269 

-0.997 

0.004* 

-25.31 % 

-97.9 % 

-0.945 

0.269 

-0.869 

0.004* 

0.007 ± 0.080 

0.008 ± 0.085  0.032 ± 0.090  0.029 ± 0.086 

28.33 % 

-8.05 % 

-1.093 

0.269 

-0.872 

0.004* 

0.017 ± 0.072 

0.014 ± 0.073  0.033 ± 0.087  0.035 ± 0.093 

-15.59 % 

4.87 % 

-0.762 

0.28 

-0.938 

0.004* 

SLF_III_left 

0.036 ± 0.081 

0.036 ± 0.082  0.062 ± 0.128  0.067 ± 0.134 

-1 % 

6.67 % 

-1.021 

0.269 

-1.169 

0.004* 

SLF_III_right 

0.047 ± 0.074 

0.049 ± 0.078  0.089 ± 0.115  0.093 ± 0.124 

5.05 % 

4.68 % 

-1.767 

0.216 

-1.739 

0.003* 

SLF_II_left 

0.027 ± 0.072 

0.028 ± 0.073  0.043 ± 0.105  0.058 ± 0.134 

4.02 % 

37.08 % 

-0.719 

0.286 

-1.236 

0.004* 

SLF_II_right 

0.035 ± 0.079 

0.033 ± 0.078  0.060 ± 0.104  0.071 ± 0.120 

-6.27 % 

17.78 % 

-1.069 

0.269 

-1.550 

0.003* 

SLF_I_left 

0.007 ± 0.069 

0.015 ± 0.067  0.025 ± 0.091  0.036 ± 0.099 

103.82 % 

45.42 % 

-0.832 

0.269 

-0.987 

0.004* 

SLF_I_right 

0.006 ± 0.074 

0.008 ± 0.072  0.026 ± 0.110  0.035 ± 0.110 

34.8 % 

36.4 % 

-0.849 

0.269 

-1.179 

0.004* 

STR_left 

-7.186e-03 ± 
7.695e-02 

0.002 ± 0.075  0.029 ± 0.096  0.041 ± 0.110 

-127.13 % 

40.41 % 

-1.598 

0.216 

-1.666 

0.003* 

STR_right 

0.006 ± 0.065 

0.010 ± 0.063  0.030 ± 0.108  0.035 ± 0.106 

63.84 % 

16.85 % 

-1.112 

0.269 

-1.200 

0.004* 

ST_FO_left 

0.010 ± 0.081 

5.806e-06 ± 
8.671e-02 

0.038 ± 0.106  0.043 ± 0.114 

-99.94 % 

14.79 % 

-1.152 

0.269 

-1.665 

0.003* 

ST_FO_right 

0.033 ± 0.080 

0.026 ± 0.081  0.079 ± 0.092  0.094 ± 0.094 

-20.48 % 

18.12 % 

-1.997 

0.216 

-2.873  <0.001 

ST_OCC_left 

0.042 ± 0.078 

0.042 ± 0.079  0.044 ± 0.097  0.062 ± 0.115 

-0.72 % 

42.71 % 

-0.071 

0.447 

-0.835 

0.004* 

ST_OCC_right 

0.045 ± 0.074 

0.044 ± 0.074  0.050 ± 0.103  0.066 ± 0.117 

-3.77 % 

30.42 % 

-0.221 

0.415 

-0.923 

0.004* 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

-1.348e-02 ± 
7.252e-02 

-6.068e-03 ± 
7.133e-02 

-1.055e-02 ± 
7.394e-02 

-4.638e-03 ± 
7.093e-02 

-2.821e-02 ± 
6.992e-02 

-1.716e-02 ± 
7.083e-02 

-1.800e-02 ± 
7.040e-02 

-1.046e-02 ± 
7.192e-02 

0.009 ± 0.103  0.018 ± 0.115 

-54.99 % 

95 % 

-1.004 

0.269 

-1.026 

0.004* 

0.010 ± 0.106  0.018 ± 0.114 

-56.06 % 

69.63 % 

-0.912 

0.269 

-0.971 

0.004* 

0.012 ± 0.102  0.024 ± 0.123 

-39.16 % 

96.42 % 

-1.852 

0.216 

-1.737 

0.003* 

0.016 ± 0.113  0.023 ± 0.124 

-41.9 % 

40.32 % 

-1.514 

0.216 

-1.400 

0.003* 

ST_PREC_left 

-6.573e-03 ± 
6.506e-02 

0.004 ± 0.066  0.049 ± 0.085  0.064 ± 0.103 

-163.74 % 

31.1 % 

-2.844 

0.095 

-2.857  <0.001 

ST_PREC_right 

0.002 ± 0.068 

0.008 ± 0.069  0.035 ± 0.096  0.044 ± 0.108 

299.6 % 

25.28 % 

-1.583 

0.216 

-1.649 

0.003* 

ST_PREF_left 

0.016 ± 0.067 

0.012 ± 0.067  0.047 ± 0.085  0.054 ± 0.100 

-24.73 % 

14.88 % 

-1.553 

0.216 

-1.995 

0.003* 

ST_PREF_right 

0.020 ± 0.065 

0.015 ± 0.065  0.060 ± 0.093  0.065 ± 0.102 

-24.18 % 

8.64 % 

-1.968 

0.216 

-2.394 

0.002* 

ST_PREM_left 

0.012 ± 0.074 

0.015 ± 0.076  0.028 ± 0.094  0.040 ± 0.118 

19.98 % 

41.08 % 

-0.732 

0.286 

-1.051 

0.004* 

ST_PREM_right 

0.016 ± 0.070 

0.018 ± 0.074  0.041 ± 0.094  0.050 ± 0.105 

12.14 % 

24.11 % 

-1.166 

0.269 

-1.428 

0.003* 

T_OCC_left 

0.038 ± 0.079 

0.039 ± 0.080  0.036 ± 0.098  0.052 ± 0.115 

1 % 

45.46 % 

0.094 

0.447 

-0.554 

0.005* 

T_OCC_right 

0.037 ± 0.075 

0.036 ± 0.075  0.043 ± 0.105  0.056 ± 0.119 

-1.82 % 

30.71 % 

-0.275 

0.408 

-0.841 

0.004* 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

-1.725e-02 ± 
7.338e-02 

-9.715e-03 ± 
7.206e-02 

0.003 ± 0.101  0.010 ± 0.111 

-43.68 % 

281.16 % 

-0.883 

0.269 

-0.849 

0.004* 

-1.994e-02 ± 
7.469e-02 

-1.372e-02 ± 
7.168e-02 

-1.538e-03 ± 
1.027e-01 

-3.074e-02 ± 
7.699e-02 

-2.048e-02 ± 
7.665e-02 

8.301e-04 ± 
1.098e-01 

0.005 ± 0.106 

-31.2 % 

-414.66 % 

-0.806 

0.269 

-0.825 

0.004* 

0.010 ± 0.126 

-33.37 % 

1118.82 % 

-1.325 

0.232 

-1.225 

0.004* 

 188 

 
 
 
 
 
Table A.13 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-
value 
P11 

t-score 
P2 

q-value 
P21 

T_POSTC_right 

T_PREC_left 

-1.552e-02 ± 
7.574e-02 

-7.507e-03 ± 
7.689e-02 

-9.899e-03 ± 
6.619e-02 

6.543e-04 ± 
6.824e-02 

0.014 ± 0.119  0.021 ± 0.126 

-51.62 % 

43.72 % 

-1.232 

0.262 

-1.123 

0.004* 

0.049 ± 0.083  0.064 ± 0.099 

-106.61 % 

31.61 % 

-2.970 

0.095 

-2.984  <0.001 

T_PREC_right 

0.003 ± 0.070 

0.009 ± 0.072  0.037 ± 0.094  0.046 ± 0.102 

196.31 % 

25.4 % 

-1.577 

0.216 

-1.650 

0.003* 

T_PREF_left 

0.014 ± 0.067 

0.010 ± 0.068  0.041 ± 0.084  0.048 ± 0.097 

-25.47 % 

16.15 % 

-1.365 

0.225 

-1.772 

0.003* 

T_PREF_right 

0.016 ± 0.066 

0.011 ± 0.067  0.050 ± 0.093  0.054 ± 0.103 

-29.73 % 

9.28 % 

-1.687 

0.216 

-2.038 

0.003* 

T_PREM_left 

0.006 ± 0.071 

0.008 ± 0.074  0.020 ± 0.090  0.030 ± 0.108 

23.69 % 

46.58 % 

-0.662 

0.301 

-0.959 

0.004* 

T_PREM_right 

0.012 ± 0.070 

0.014 ± 0.074  0.033 ± 0.089  0.044 ± 0.096 

13.12 % 

32.52 % 

-0.997 

0.269 

-1.352 

0.004* 

UF_left 

0.025 ± 0.077 

0.016 ± 0.079  0.047 ± 0.112  0.057 ± 0.116 

-33.31 % 

21.04 % 

-0.928 

0.269 

-1.626 

0.003* 

0.034 ± 0.071 

0.033 ± 0.066  0.074 ± 0.086  0.087 ± 0.096 

UF_right 
Tables A.13- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

17.82 % 

-3.78 % 

-1.898 

0.216 

-2.634  <0.001 

 189 

 
 
 
 
 
 
 
 
 
Table A.14 Results of Post Hoc Tract Specific Comparisons of Fixel-Based Analysis Fiber Density 
Cross Section 

TractSeg 
Abbv. 

AF_left 

AF_right 

ATR_left 

mTBI 

Controls 

Percent Change 

P1 

P2 

P1 

P2 

mTBI 

Controls 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

0.302 ± 0.033 

0.302 ± 0.029 

0.322 ± 0.045 

0.327 ± 0.043 

0.27 % 

1.51 % 

-2.085 

0.001* 

-2.690  <0.001 

0.314 ± 0.031 

0.312 ± 0.031 

0.327 ± 0.044 

0.333 ± 0.046 

-0.41 % 

1.69 % 

-1.382 

0.004* 

-2.081  <0.001 

0.304 ± 0.032 

0.301 ± 0.033 

0.332 ± 0.044 

0.335 ± 0.046 

-0.96 % 

1.09 % 

-2.860  <0.001 

-3.420  <0.001 

ATR_right 

0.300 ± 0.030 

0.299 ± 0.031 

0.328 ± 0.036 

0.335 ± 0.042 

-0.38 % 

2.22 % 

-3.160  <0.001 

-3.855  <0.001 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

0.274 ± 0.040 

0.277 ± 0.037 

0.316 ± 0.047 

0.329 ± 0.040 

1.14 % 

4.09 % 

-3.595  <0.001 

-4.861  <0.001 

0.388 ± 0.051 

0.380 ± 0.051 

0.435 ± 0.068 

0.437 ± 0.058 

-2.08 % 

0.44 % 

-3.058  <0.001 

-3.878  <0.001 

0.342 ± 0.031 

0.340 ± 0.032 

0.370 ± 0.045 

0.375 ± 0.043 

-0.68 % 

1.34 % 

-2.961  <0.001 

-3.653  <0.001 

0.337 ± 0.032 

0.340 ± 0.036 

0.363 ± 0.053 

0.370 ± 0.056 

0.84 % 

2.06 % 

-2.471  <0.001 

-2.668  <0.001 

0.371 ± 0.033 

0.375 ± 0.034 

0.408 ± 0.038 

0.420 ± 0.040 

1.1 % 

2.77 % 

-3.890  <0.001 

-4.429  <0.001 

0.368 ± 0.032 

0.371 ± 0.034 

0.392 ± 0.044 

0.400 ± 0.041 

0.81 % 

2.01 % 

-2.509  <0.001 

-2.977  <0.001 

0.360 ± 0.042 

0.360 ± 0.036 

0.369 ± 0.046 

0.370 ± 0.042 

0.12 % 

0.19 % 

-0.803 

0.008* 

-0.955 

0.005* 

0.353 ± 0.046 

0.353 ± 0.039 

0.353 ± 0.042 

0.359 ± 0.043 

-0.03 % 

1.61 % 

-0.051 

0.015* 

-0.575 

0.008* 

0.352 ± 0.030 

0.352 ± 0.029 

0.371 ± 0.041 

0.376 ± 0.038 

-0.04 % 

1.25 % 

-2.052 

0.001* 

-2.718  <0.001 

0.322 ± 0.030 

0.320 ± 0.030 

0.341 ± 0.051 

0.340 ± 0.036 

-0.63 % 

-0.14 % 

-1.886 

0.002* 

-2.264  <0.001 

0.322 ± 0.031 

0.319 ± 0.032 

0.333 ± 0.048 

0.336 ± 0.038 

-0.8 % 

0.91 % 

-1.159 

0.005* 

-1.837 

0.001* 

0.424 ± 0.042 

0.431 ± 0.040 

0.472 ± 0.049 

0.483 ± 0.047 

1.66 % 

2.37 % 

-3.882  <0.001 

-4.450  <0.001 

CST_right 

0.440 ± 0.041 

0.443 ± 0.039 

0.473 ± 0.052 

0.480 ± 0.050 

0.71 % 

1.6 % 

-2.697  <0.001 

-3.201  <0.001 

FPT_left 

0.404 ± 0.037 

0.406 ± 0.036 

0.435 ± 0.042 

0.443 ± 0.042 

0.64 % 

1.96 % 

-2.872  <0.001 

-3.505  <0.001 

FPT_right 

0.404 ± 0.034 

0.405 ± 0.033 

0.432 ± 0.048 

0.439 ± 0.045 

0.21 % 

1.63 % 

-2.595  <0.001 

-3.290  <0.001 

FX_left 

FX_right 

ICP_left 

0.190 ± 0.044 

0.192 ± 0.049 

0.207 ± 0.046 

0.211 ± 0.051 

0.98 % 

2.17 % 

-1.362 

0.004* 

-1.427 

0.002* 

0.199 ± 0.045 

0.200 ± 0.049 

0.206 ± 0.050 

0.211 ± 0.050 

0.56 % 

2.76 % 

-0.513 

0.01* 

-0.816 

0.006* 

0.315 ± 0.040 

0.315 ± 0.041 

0.321 ± 0.034 

0.323 ± 0.029 

0.04 % 

0.66 % 

-0.550 

0.01* 

-0.731 

0.006* 

ICP_right 

0.287 ± 0.031 

0.285 ± 0.032 

0.290 ± 0.031 

0.291 ± 0.030 

-0.65 % 

0.57 % 

-0.302 

0.012* 

-0.686 

0.007* 

IFO_left 

0.328 ± 0.032 

0.326 ± 0.032 

0.342 ± 0.042 

0.347 ± 0.042 

-0.44 % 

1.32 % 

-1.560 

0.003* 

-2.161  <0.001 

IFO_right 

0.332 ± 0.030 

0.330 ± 0.029 

0.345 ± 0.039 

0.351 ± 0.042 

-0.5 % 

1.75 % 

-1.531 

0.003* 

-2.420  <0.001 

ILF_left 

0.311 ± 0.038 

0.313 ± 0.035 

0.320 ± 0.044 

0.327 ± 0.040 

0.64 % 

2.31 % 

-0.774 

0.008* 

-1.356 

0.003* 

ILF_right 

0.322 ± 0.035 

0.322 ± 0.035 

0.329 ± 0.042 

0.337 ± 0.036 

-0.01 % 

2.28 % 

-0.686 

0.009* 

-1.484 

0.002* 

MCP 

0.330 ± 0.040 

0.330 ± 0.040 

0.333 ± 0.039 

0.336 ± 0.037 

-0.03 % 

0.82 % 

-0.272 

0.012* 

-0.535 

0.008* 

MLF_left 

0.319 ± 0.037 

0.320 ± 0.031 

0.330 ± 0.044 

0.330 ± 0.037 

0.37 % 

0.15 % 

-1.014 

0.006* 

-1.125 

0.004* 

MLF_right 

0.337 ± 0.036 

0.337 ± 0.033 

0.342 ± 0.040 

0.342 ± 0.037 

-0.06 % 

-0.06 % 

-0.453 

0.01* 

-0.500 

0.008* 

OR_left 

0.330 ± 0.037 

0.332 ± 0.038 

0.340 ± 0.037 

0.350 ± 0.040 

0.59 % 

2.85 % 

-0.982 

0.006* 

-1.663 

0.002* 

OR_right 

0.350 ± 0.037 

0.351 ± 0.035 

0.356 ± 0.040 

0.364 ± 0.045 

0.26 % 

2.17 % 

-0.604 

0.009* 

-1.268 

0.003* 

POPT_left 

0.373 ± 0.036 

0.376 ± 0.033 

0.394 ± 0.046 

0.398 ± 0.042 

0.95 % 

1 % 

-2.006 

0.001* 

-2.228  <0.001 

POPT_right 

0.398 ± 0.039 

0.400 ± 0.036 

0.418 ± 0.051 

0.421 ± 0.050 

0.58 % 

0.76 % 

-1.669 

0.003* 

-1.861 

0.001* 

SCP_left 

0.348 ± 0.039 

0.350 ± 0.041 

0.364 ± 0.038 

0.370 ± 0.033 

0.71 % 

1.61 % 

-1.516 

0.003* 

-1.767 

0.002* 

SCP_right 

0.349 ± 0.032 

0.349 ± 0.032 

0.360 ± 0.039 

0.365 ± 0.035 

0.02 % 

1.44 % 

-1.155 

0.005* 

-1.745 

0.002* 

SLF_III_left 

0.322 ± 0.037 

0.322 ± 0.035 

0.344 ± 0.063 

0.348 ± 0.061 

0.03 % 

1.08 % 

-1.814 

0.002* 

-2.169  <0.001 

SLF_III_right 

0.321 ± 0.038 

0.320 ± 0.038 

0.334 ± 0.055 

0.337 ± 0.054 

-0.2 % 

1.05 % 

-1.097 

0.006* 

-1.456 

0.002* 

SLF_II_left 

0.298 ± 0.032 

0.299 ± 0.032 

0.309 ± 0.045 

0.316 ± 0.052 

0.13 % 

2.52 % 

-1.050 

0.006* 

-1.714 

0.002* 

SLF_II_right 

0.298 ± 0.036 

0.297 ± 0.035 

0.309 ± 0.042 

0.314 ± 0.045 

-0.54 % 

1.47 % 

-1.024 

0.006* 

-1.599 

0.002* 

SLF_I_left 

0.283 ± 0.030 

0.284 ± 0.029 

0.291 ± 0.036 

0.294 ± 0.032 

0.45 % 

1.04 % 

-0.905 

0.007* 

-1.152 

0.004* 

SLF_I_right 

0.297 ± 0.033 

0.297 ± 0.031 

0.309 ± 0.043 

0.311 ± 0.040 

0.12 % 

0.82 % 

-1.242 

0.005* 

-1.545 

0.002* 

STR_left 

0.416 ± 0.041 

0.423 ± 0.044 

0.453 ± 0.052 

0.467 ± 0.055 

1.72 % 

3.07 % 

-3.069  <0.001 

-3.411  <0.001 

STR_right 

0.439 ± 0.034 

0.441 ± 0.036 

0.466 ± 0.057 

0.474 ± 0.056 

0.45 % 

1.69 % 

-2.394  <0.001 

-2.872  <0.001 

ST_FO_left 

0.327 ± 0.039 

0.323 ± 0.039 

0.358 ± 0.053 

0.362 ± 0.050 

-1.24 % 

1.27 % 

-2.621  <0.001 

-3.373  <0.001 

ST_FO_right 

0.351 ± 0.040 

0.347 ± 0.038 

0.387 ± 0.048 

0.397 ± 0.048 

-1.32 % 

2.76 % 

-3.002  <0.001 

-4.414  <0.001 

ST_OCC_left 

0.341 ± 0.037 

0.343 ± 0.037 

0.353 ± 0.036 

0.363 ± 0.039 

0.48 % 

2.75 % 

-1.114 

0.006* 

-1.857 

0.001* 

 190 

 
 
 
 
 
 
Table A.14 (cont’d)  

TractSeg Abbv. 

P1 

P2 

P1 

P2 

mTBI 

Controls 

mTBI 

Controls 

Percent Change 

t-score 
P1 

q-value 
P11 

t-score 
P2 

q-value 
P21 

ST_OCC_right 

0.351 ± 0.036  0.352 ± 0.033  0.360 ± 0.041  0.367 ± 0.045 

0.19 % 

2 % 

-0.841 

0.007* 

-1.523 

0.002* 

ST_PAR_left 

0.334 ± 0.032  0.337 ± 0.031  0.354 ± 0.043  0.358 ± 0.040 

0.84 % 

1.01 % 

-2.068 

0.001* 

-2.224  <0.001 

ST_PAR_right 

0.357 ± 0.035  0.357 ± 0.032  0.372 ± 0.045  0.375 ± 0.045 

0.25 % 

0.86 % 

-1.467 

0.004* 

-1.784 

0.002* 

ST_POSTC_left 

0.330 ± 0.030  0.336 ± 0.033  0.364 ± 0.043  0.372 ± 0.041 

1.88 % 

2.18 % 

-3.780  <0.001 

-3.753  <0.001 

ST_POSTC_right  0.357 ± 0.031  0.359 ± 0.031  0.385 ± 0.047  0.390 ± 0.047 

0.67 % 

1.27 % 

-2.878  <0.001 

-3.151  <0.001 

ST_PREC_left 

0.346 ± 0.029  0.352 ± 0.031  0.388 ± 0.039  0.399 ± 0.038 

1.79 % 

2.66 % 

-4.821  <0.001 

-5.047  <0.001 

ST_PREC_right 

0.367 ± 0.030  0.370 ± 0.030  0.395 ± 0.043  0.402 ± 0.044 

0.73 % 

1.82 % 

-3.022  <0.001 

-3.478  <0.001 

ST_PREF_left 

0.339 ± 0.030  0.338 ± 0.031  0.367 ± 0.039  0.372 ± 0.040 

-0.25 % 

1.53 % 

-3.085  <0.001 

-3.699  <0.001 

ST_PREF_right 

0.340 ± 0.029  0.339 ± 0.028  0.367 ± 0.041  0.374 ± 0.043 

-0.56 % 

1.96 % 

-2.970  <0.001 

-3.968  <0.001 

ST_PREM_left 

0.316 ± 0.030  0.318 ± 0.030  0.336 ± 0.044  0.344 ± 0.048 

0.5 % 

2.17 % 

-2.173 

0.001* 

-2.665  <0.001 

ST_PREM_right 

0.331 ± 0.031  0.334 ± 0.032  0.357 ± 0.041  0.363 ± 0.040 

0.78 % 

1.85 % 

-2.718  <0.001 

-3.133  <0.001 

T_OCC_left 

0.324 ± 0.036  0.326 ± 0.037  0.334 ± 0.035  0.344 ± 0.039 

0.54 % 

2.92 % 

-0.940 

0.007* 

-1.674 

0.002* 

T_OCC_right 

0.342 ± 0.037  0.343 ± 0.034  0.349 ± 0.040  0.356 ± 0.045 

0.28 % 

2.03 % 

-0.665 

0.009* 

-1.282 

0.003* 

T_PAR_left 

0.334 ± 0.032  0.336 ± 0.031  0.352 ± 0.042  0.356 ± 0.041 

0.75 % 

1.06 % 

-1.881 

0.002* 

-2.084  <0.001 

T_PAR_right 

0.361 ± 0.036  0.362 ± 0.033  0.374 ± 0.044  0.378 ± 0.045 

0.34 % 

0.98 % 

-1.294 

0.005* 

-1.610 

0.002* 

T_POSTC_left 

0.347 ± 0.034  0.353 ± 0.036  0.377 ± 0.047  0.385 ± 0.048 

1.52 % 

2.14 % 

-2.816  <0.001 

-2.964  <0.001 

T_POSTC_right 

0.368 ± 0.034  0.371 ± 0.034  0.395 ± 0.052  0.401 ± 0.052 

0.64 % 

1.47 % 

-2.469  <0.001 

-2.811  <0.001 

T_PREC_left 

0.365 ± 0.033  0.371 ± 0.034  0.411 ± 0.041  0.422 ± 0.043 

1.71 % 

2.83 % 

-4.691  <0.001 

-4.997  <0.001 

T_PREC_right 

0.382 ± 0.032  0.385 ± 0.032  0.411 ± 0.045  0.419 ± 0.046 

0.75 % 

2.02 % 

-2.916  <0.001 

-3.423  <0.001 

T_PREF_left 

0.342 ± 0.031  0.342 ± 0.031  0.371 ± 0.039  0.377 ± 0.041 

-0.08 % 

1.71 % 

-3.189  <0.001 

-3.783  <0.001 

T_PREF_right 

0.331 ± 0.028  0.330 ± 0.028  0.357 ± 0.039  0.364 ± 0.042 

-0.37 % 

1.97 % 

-3.038  <0.001 

-3.861  <0.001 

T_PREM_left 

0.331 ± 0.031  0.333 ± 0.032  0.356 ± 0.043  0.363 ± 0.047 

0.57 % 

2.05 % 

-2.648  <0.001 

-3.042  <0.001 

T_PREM_right 

0.344 ± 0.032  0.347 ± 0.034  0.373 ± 0.041  0.381 ± 0.042 

0.88 % 

2.23 % 

-3.029  <0.001 

-3.411  <0.001 

UF_left 

0.311 ± 0.036  0.307 ± 0.035  0.334 ± 0.051  0.334 ± 0.047 

-1.09 % 

0.18 % 

-2.039 

0.001* 

-2.519  <0.001 

0.308 ± 0.031  0.305 ± 0.031  0.335 ± 0.035  0.341 ± 0.040 

UF_right 
Tables A.14- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

-2.984  <0.001 

-0.79 % 

1.88 % 

-3.842  <0.001 

 191 

 
 
 
 
 
 
 
 
 
Figure A.1 – Boxplots for t-test results in the Arcuate Fascicle (Left) 

 192 

 
Figure A.2- Boxplots for t-test results in the Arcuate Fascicle (Left) 

 193 

 
 
Figure A.3- Boxplots for t-test results in the Arcuate Fascicle (right) 

 194 

 
 
Figure A.4- Boxplots for t-test results in the Anterior Thalamic Radiation (left) 

 195 

 
 
Figure A.5- Boxplots for t-test results in the Anterior Thalamic Radiation (left) 

 196 

 
 
Figure A.6- Boxplots for t-test results in the Anterior Thalamic Radiation (Right) 

 197 

 
 
Figure A.7- Boxplots for t-test results in the Anterior Commissure 

 198 

 
 
 
Figure A.8-  Boxplots for t-test results in the Anterior Commissure 

 199 

 
 
 
 
Figure A.9- Boxplots for t-test results in the Corpus Callosum-Rostrum 

 200 

 
 
 
Figure A.10- Boxplots for t-test results in the Corpus Callosum - Genu 

 201 

 
 
 
 
Figure A.11- Boxplots for t-test results in the Corpus Callosum - Genu 

 202 

 
 
Figure A.12- Boxplots for t-test results in the Corpus Callosum – Rostral Body Premotor 

 203 

 
 
 
 
 
 
Figure A.13- Boxplots for t-test results in the Corpus Callosum – Rostral 

 204 

 
 
Figure A.14- Boxplots for t-test results in the Corpus Callosum - Anterior midbody (Primary Motor) 

 205 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure A.15- Boxplots for t-test results in the  Corpus Callosum - Posterior midbody (Primary 
Somatosensory) 

 206 

 
 
Figure A.16- Boxplots for t-test results in the Corpus Callosum - Isthmus 

 207 

 
 
 
 
 
Figure A.17- Boxplots for t-test results in the Corpus Callosum - Splenium 

 208 

 
 
 
 
Figure A.18- Boxplots for t-test results in the Cingulum (left) 

 209 

 
 
 
 
Figure A.19- Boxplots for t-test results in the Cingulum (right) 

 210 

 
 
 
 
Figure A.20- Boxplots for t-test results in the Corticospinal Tract (Left) 

 211 

 
 
Figure A.22- Boxplots for t-test results in the Corticospinal Tract (Right) 

 212 

 
 
 
 
 
Figure A.23 - Boxplots for t-test results in the Corticospinal tract (right) 

 213 

 
 
 
Figure A.24- Boxplots for t-test results in the Fronto-pontine tract (left) 

 214 

 
 
Figure A.25- Boxplots for t-test results in the Fronto-pontine tract (left) 

 215 

 
 
Figure A.26- Boxplots for t-test results in the Fronto-pontine tract (right) 

 216 

 
 
 
 
Figure A.27- Boxplots for t-test results in the Frontopontine Tract (Right) 

 217 

 
 
Figure A.28- Boxplots for t-test results in the Inferior cerebellar peduncle (left) 

 218 

 
 
 
Figure A.29- Boxplots for t-test results in the Inferior cerebellar peduncle (right) 

 219 

 
 
Figure A.30- Boxplots for t-test results in the Inferior occipito-frontal fascicle (left) 

 220 

 
 
Figure A.31- Boxplots for t-test results in the Inferior occipito-frontal fascicle (left) 

 221 

 
 
 
Figure A.32- Boxplots for t-test results in the Inferior occipito-frontal fascicle (right) 

 222 

 
 
Figure A.33-  Boxplots for t-test results in the Inferior longitudinal fascicle (left) 

 223 

 
 
 
 
 
Figure A.34- Boxplots for t-test results in the Inferior longitudinal fascicle (right) 

 224 

 
 
 
 
Figure A.35- Boxplots for t-test results in the Middle cerebellar peduncle 

 225 

 
 
 
 
 
Figure A.36- Boxplots for t-test results in the Middle longitudinal fascicle (left) 

 226 

 
 
 
 
Figure A.37- Boxplots for t-test results in the Middle longitudinal fascicle (right) 

 227 

 
 
 
 
Figure A.38- Boxplots for t-test results in the Optic radiation (left) 

 228 

 
 
 
 
Figure A.39- Boxplots for t-test results in the Optic radiation (right) 

 229 

 
 
 
 
Figure A.40- Boxplots for t-test results in the Parieto-occipital pontine (left) 

 230 

 
 
 
 
Figure A.41- Boxplots for t-test results in the Parieto-occipital pontine (right) 

 231 

 
 
 
 
Figure A.42- Boxplots for t-test results in the Superior cerebellar peduncle (left) 

 232 

 
 
 
 
Figure A.43- Boxplots for t-test results in the Superior cerebellar peduncle (left) 

 233 

 
 
Figure A.44- Boxplots for t-test results in the Superior cerebellar peduncle (right) 

 234 

 
 
 
Figure A.45- Boxplots for t-test results in the Superior longitudinal fascicle I (left) 

 235 

 
 
 
 
Figure A.46- Boxplots for t-test results in the Superior longitudinal fascicle I (right) 

 236 

 
 
 
 
Figure A.47- Boxplots for t-test results in the Superior longitudinal fascicle II (left) 

 237 

 
 
 
 
Figure A.48- Boxplots for t-test results in the Superior longitudinal fascicle II (right) 

 238 

 
 
 
 
Figure A.49- Boxplots for t-test results in the Superior longitudinal fascicle III (left) 

 239 

 
 
 
 
Figure A.50- Boxplots for t-test results in the Superior longitudinal fascicle III (left) 

 240 

 
 
 
 
Figure A.51- Boxplots for t-test results in the Superior longitudinal fascicle III (right) 

 241 

 
 
 
 
Figure A.52- Boxplots for t-test results in the Superior Thalamic Radiation (left) 

 242 

 
 
Figure A.53- Boxplots for t-test results in the Superior Thalamic Radiation (left) 

 243 

 
 
Figure A.54- Boxplots for t-test results in the Superior Thalamic Radiation (right) 

 244 

 
 
 
 
 
 
Figure A.55- Boxplots for t-test results in the Superior Thalamic Radiation (right) 

 245 

 
 
 
 
Figure A.56- Boxplots for t-test results in the Striato-fronto-orbital (left) 

 246 

 
 
Figure A.57- Boxplots for t-test results in the Striato-fronto-orbital (left) 

 247 

 
 
 
 
 
Figure A.58- Boxplots for t-test results in the Striato-fronto-orbital (right) 

 248 

 
 
 
 
Figure A.59- Boxplots for t-test results in the Striato-occipital (left) 

 249 

 
 
 
 
Figure A.60- Boxplots for t-test results in the Striato-occipital (left) 

 250 

 
 
 
 
Figure A.61- Boxplots for t-test results in the Striato-occipital (right) 

 251 

 
 
 
 
Figure A.62- Boxplots for t-test results in the Striato-parietal (left) 

 252 

 
 
 
 
Figure A.63- Boxplots for t-test results in the Striato-parietal (right) 

 253 

 
 
 
 
Figure A.64- Boxplots for t-test results in the Striato-postcentral (left) 

 254 

 
 
 
Figure A.65- Boxplots for t-test results in the Striato-postcentral (right) 

 255 

 
 
 
Figure A.66- Boxplots for t-test results in the Striato-precentral (left) 

 256 

 
 
 
Figure A.67- Boxplots for t-test results in the Striato-precentral (left) 

 257 

 
 
Figure A.68- Boxplots for t-test results in the Striato-precentral (right) 

 258 

 
 
 
 
 
 
 
Figure A.69- Boxplots for t-test results in the Striato-prefrontal (left) 

 259 

 
 
Figure A.70- Boxplots for t-test results in the Striato-prefrontal (left) 

 260 

 
 
Figure A.71- Boxplots for t-test results in the Striato-prefrontal (right) 

 261 

 
 
 
 
 
Figure A.72- Boxplots for t-test results in the Striato-prefrontal (right) 

 262 

 
 
Figure A.73- Boxplots for t-test results in the Striato-premotor (left) 

 263 

 
 
Figure A.74- Boxplots for t-test results in the Striato-premotor (left) 

 264 

 
 
Figure A.75-  Boxplots for t-test results in the Striato-premotor (right) 

 265 

 
 
 
 
 
Figure A.76- Boxplots for t-test results in the Striato-premotor (right) 

 266 

 
 
Figure A.77- Boxplots for t-test results in the Thalamo-occipital (left) 

 267 

 
 
 
 
 
Figure A.78- Boxplots for t-test results in the Thalamo-occipital (right) 

 268 

 
 
 
 
Figure A.79- Boxplots for t-test results in the Thalamo-parietal (left) 

 269 

 
 
 
 
Figure A.80- Boxplots for t-test results in the Thalamo-parietal (right) 

 270 

 
 
 
 
Figure A.81- Boxplots for t-test results in the Thalamo-postcentral (left) 

 271 

 
 
 
 
Figure A.82- Boxplots for t-test results in the Thalamo-postcentral (right) 

 272 

 
 
 
 
Figure A.83- Boxplots for t-test results in the Thalamo-precentral (left) 

 273 

 
 
 
 
Figure A.84- Boxplots for t-test results in the Thalamo-precentral (left) 

 274 

 
 
Figure A.85- Boxplots for t-test results in the Thalamo-precentral (right) 

 275 

 
 
 
 
Figure A.86- Boxplots for t-test results in the Thalamo-prefrontal (left) 

 276 

 
 
 
 
Figure A.87- Boxplots for t-test results in the Thalamo-prefrontal (left) 

 277 

 
 
Figure A.88- Boxplots for t-test results in the Thalamo-prefrontal (right) 

 278 

 
 
 
 
 
Figure A.89 - Boxplots for t-test results in the Thalamo-premotor (left) 

 279 

 
 
 
Figure A.90- Boxplots for t-test results in the Thalamo-premotor (left) 

 280 

 
 
Figure A.91- Boxplots for t-test results in the Thalamo-premotor (right) 

 281 

 
 
 
 
Figure A.92- Boxplots for t-test results in the Uncinate fascicle (left) 

 282 

 
 
 
 
Figure A.93- Boxplots for t-test results in the Uncinate fascicle (left) 

 283 

 
 
 
Figure A.94- Boxplots for t-test results in the Uncinate fascicle (right) 

 284 

 
 
Table A.15 Results of Post Hoc Tract Specific Comparison of Diffusion Tensor Metric Fractional 
Anisotropy 

mTBI 

Controls 

TractSeg Name 

P1 

AF_left 

0.256 ± 0.013 

AF_right 

0.267 ± 0.011 

ATR_left 

0.271 ± 0.011 

ATR_right 

0.263 ± 0.010 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

0.270 ± 0.017 

0.343 ± 0.023 

0.296 ± 0.014 

0.291 ± 0.017 

0.307 ± 0.018 

0.318 ± 0.017 

0.301 ± 0.016 

0.308 ± 0.015 

0.298 ± 0.013 

CG_left 

0.278 ± 0.017 

CG_right 

0.282 ± 0.019 

CST_left 

0.344 ± 0.014 

CST_right 

0.352 ± 0.013 

FPT_left 

0.333 ± 0.014 

FPT_right 

0.333 ± 0.013 

FX_left 

0.288 ± 0.046 

FX_right 

0.294 ± 0.045 

ICP_left 

0.275 ± 0.025 

ICP_right 

0.258 ± 0.016 

IFO_left 

0.283 ± 0.011 

IFO_right 

0.281 ± 0.010 

ILF_left 

0.277 ± 0.011 

ILF_right 

0.273 ± 0.011 

P2 

0.256 ± 
0.011 

0.266 ± 
0.013 

0.271 ± 
0.012 

0.264 ± 
0.012 

0.267 ± 
0.022 

0.336 ± 
0.032 

0.295 ± 
0.013 

0.290 ± 
0.021 

0.305 ± 
0.019 

0.314 ± 
0.020 

0.300 ± 
0.015 

0.306 ± 
0.016 

0.297 ± 
0.013 

0.281 ± 
0.021 

0.283 ± 
0.016 

0.345 ± 
0.016 

0.351 ± 
0.017 

0.335 ± 
0.015 

0.333 ± 
0.014 

0.281 ± 
0.045 

0.289 ± 
0.049 

0.273 ± 
0.030 

0.255 ± 
0.017 

0.281 ± 
0.011 

0.279 ± 
0.012 

0.275 ± 
0.017 

0.272 ± 
0.019 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

P1 

P2 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

0.259 ± 
0.019 

0.263 ± 
0.016 

0.278 ± 
0.020 

0.266 ± 
0.016 

0.287 ± 
0.024 

0.351 ± 
0.023 

0.298 ± 
0.023 

0.295 ± 
0.035 

0.305 ± 
0.032 

0.313 ± 
0.033 

0.295 ± 
0.023 

0.304 ± 
0.014 

0.296 ± 
0.022 

0.279 ± 
0.023 

0.280 ± 
0.025 

0.347 ± 
0.025 

0.350 ± 
0.025 

0.334 ± 
0.027 

0.331 ± 
0.023 

0.295 ± 
0.047 

0.294 ± 
0.048 

0.260 ± 
0.029 

0.242 ± 
0.025 

0.285 ± 
0.013 

0.281 ± 
0.011 

0.283 ± 
0.012 

0.278 ± 
0.013 

0.260 ± 
0.016 

0.261 ± 
0.026 

0.280 ± 
0.019 

0.269 ± 
0.018 

0.286 ± 
0.030 

0.343 ± 
0.056 

0.298 ± 
0.025 

0.295 ± 
0.034 

0.307 ± 
0.031 

0.315 ± 
0.035 

0.293 ± 
0.026 

0.297 ± 
0.037 

0.294 ± 
0.025 

0.277 ± 
0.027 

0.277 ± 
0.025 

0.351 ± 
0.024 

0.350 ± 
0.031 

0.336 ± 
0.027 

0.335 ± 
0.025 

0.291 ± 
0.055 

0.300 ± 
0.052 

0.263 ± 
0.044 

0.248 ± 
0.031 

0.279 ± 
0.029 

0.279 ± 
0.027 

0.274 ± 
0.032 

0.272 ± 
0.033 

-0.12 % 

0.41 % 

0.189 

0.456 

-0.319 

0.995 

-0.45 % 

-0.64 % 

0.958 

0.399 

0.181 

0.995 

-0.03 % 

0.74 % 

0.037 

0.49 

-0.507 

0.995 

0.3 % 

1.3 % 

-0.466 

0.414 

-0.760 

0.995 

-1.11 % 

-0.4 % 

1.164 

0.399 

0.125 

0.995 

-1.88 % 

-2.17 % 

1.895 

0.399 

0.443 

0.995 

-0.27 % 

-0.03 % 

0.663 

0.402 

-0.150 

0.995 

-0.47 % 

0.16 % 

0.511 

0.414 

-0.199 

0.995 

-0.63 % 

0.83 % 

0.857 

0.399 

-0.380 

0.995 

-1.12 % 

0.6 % 

1.445 

0.399 

-0.239 

0.995 

-0.25 % 

-0.57 % 

0.513 

0.414 

0.167 

0.995 

-0.57 % 

-2.3 % 

1.229 

0.399 

0.742 

0.995 

-0.27 % 

-0.43 % 

0.669 

0.402 

0.061 

0.995 

1.02 % 

-0.92 % 

-0.698 

0.402 

0.253 

0.995 

0.27 % 

-1 % 

-0.304 

0.435 

0.059 

0.995 

0.46 % 

1.16 % 

-0.848 

0.399 

-0.557 

0.995 

-0.08 % 

-0.01 % 

0.287 

0.435 

0.052 

0.995 

0.52 % 

0.65 % 

-0.715 

0.402 

-0.259 

0.995 

0 % 

1.01 % 

0.192 

0.456 

-0.445 

0.995 

-2.34 % 

-1.51 % 

1.431 

0.399 

0.181 

0.995 

-1.54 % 

1.95 % 

1.048 

0.399 

-1.168 

0.995 

-0.62 % 

1.02 % 

1.003 

0.399 

-0.202 

0.995 

-1.4 % 

2.42 % 

2.396 

0.399 

-0.555 

0.995 

-0.51 % 

-1.94 % 

1.292 

0.399 

0.664 

0.995 

-0.47 % 

-0.63 % 

1.460 

0.399 

0.340 

0.995 

-0.7 % 

-3.15 % 

1.039 

0.399 

1.151 

0.995 

-0.66 % 

-2.31 % 

1.266 

0.399 

0.730 

0.995 

 285 

 
 
 
 
 
 
 
 
Table A.15 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_lef
t 

ST_POSTC_rig
ht 

ST_PREC_left 

ST_PREC_righ
t 

0.301 ± 
0.015 

0.265 ± 
0.014 

0.272 ± 
0.013 

0.293 ± 
0.014 

0.300 ± 
0.013 

0.313 ± 
0.013 

0.327 ± 
0.012 

0.313 ± 
0.019 

0.313 ± 
0.013 

0.264 ± 
0.015 

0.273 ± 
0.012 

0.250 ± 
0.015 

0.252 ± 
0.014 

0.243 ± 
0.014 

0.257 ± 
0.014 

0.335 ± 
0.017 

0.339 ± 
0.016 

0.285 ± 
0.017 

0.290 ± 
0.017 

0.294 ± 
0.014 

0.295 ± 
0.013 

0.280 ± 
0.012 

0.296 ± 
0.012 

0.291 ± 
0.014 

0.308 ± 
0.014 

0.295 ± 
0.012 

0.305 ± 
0.012 

0.299 ± 
0.016 

0.266 ± 
0.014 

0.271 ± 
0.014 

0.292 ± 
0.014 

0.300 ± 
0.015 

0.313 ± 
0.015 

0.326 ± 
0.016 

0.313 ± 
0.023 

0.311 ± 
0.015 

0.262 ± 
0.016 

0.270 ± 
0.020 

0.253 ± 
0.016 

0.253 ± 
0.016 

0.245 ± 
0.018 

0.259 ± 
0.019 

0.336 ± 
0.019 

0.339 ± 
0.020 

0.281 ± 
0.018 

0.288 ± 
0.020 

0.293 ± 
0.013 

0.294 ± 
0.014 

0.282 ± 
0.016 

0.295 ± 
0.015 

0.292 ± 
0.018 

0.306 ± 
0.020 

0.296 ± 
0.015 

0.305 ± 
0.017 

0.291 ± 
0.029 

0.262 ± 
0.018 

0.266 ± 
0.020 

0.294 ± 
0.009 

0.300 ± 
0.015 

0.309 ± 
0.022 

0.323 ± 
0.023 

0.308 ± 
0.026 

0.302 ± 
0.029 

0.263 ± 
0.028 

0.263 ± 
0.020 

0.249 ± 
0.022 

0.247 ± 
0.021 

0.238 ± 
0.028 

0.252 ± 
0.027 

0.338 ± 
0.024 

0.339 ± 
0.027 

0.295 ± 
0.016 

0.294 ± 
0.016 

0.294 ± 
0.013 

0.295 ± 
0.015 

0.278 ± 
0.019 

0.292 ± 
0.019 

0.292 ± 
0.019 

0.307 ± 
0.022 

0.300 ± 
0.019 

0.305 ± 
0.019 

0.291 ± 
0.037 

0.264 ± 
0.016 

0.264 ± 
0.023 

0.289 ± 
0.035 

0.297 ± 
0.026 

0.311 ± 
0.022 

0.325 ± 
0.024 

0.315 ± 
0.034 

0.310 ± 
0.028 

0.266 ± 
0.022 

0.259 ± 
0.038 

0.255 ± 
0.027 

0.249 ± 
0.018 

0.239 ± 
0.023 

0.256 ± 
0.022 

0.338 ± 
0.035 

0.344 ± 
0.023 

0.295 ± 
0.033 

0.293 ± 
0.037 

0.289 ± 
0.036 

0.292 ± 
0.024 

0.279 ± 
0.018 

0.293 ± 
0.020 

0.296 ± 
0.015 

0.308 ± 
0.023 

0.301 ± 
0.017 

0.307 ± 
0.019 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.87 % 

0.02 % 

1.749 

0.399 

0.026 

0.995 

0.39 % 

0.57 % 

-0.431 

0.414 

-0.357 

0.995 

-0.35 % 

-0.71 % 

0.718 

0.402 

0.166 

0.995 

-0.43 % 

-1.73 % 

0.983 

0.399 

0.683 

0.995 

-0.26 % 

-0.99 % 

0.895 

0.399 

0.504 

0.995 

0.14 % 

0.84 % 

-0.078 

0.488 

-0.345 

0.995 

-0.16 % 

0.52 % 

0.463 

0.414 

-0.104 

0.995 

-0.01 % 

2.26 % 

0.379 

0.423 

-0.644 

0.995 

-0.66 % 

2.62 % 

1.247 

0.399 

-0.736 

0.995 

-0.76 % 

1.03 % 

0.931 

0.399 

-0.472 

0.995 

-1.15 % 

-1.43 % 

1.294 

0.399 

0.351 

0.995 

1.16 % 

2.16 % 

-1.260 

0.399 

-0.826 

0.995 

0.5 % 

0.96 % 

-0.359 

0.424 

-0.463 

0.995 

1 % 

0.68 % 

-1.019 

0.399 

-0.287 

0.995 

0.97 % 

1.65 % 

-0.889 

0.399 

-0.540 

0.995 

0.47 % 

0.18 % 

-0.636 

0.402 

-0.007 

0.995 

-0.13 % 

1.42 % 

0.021 

0.49 

-0.937 

0.995 

-1.41 % 

0.09 % 

1.948 

0.399 

0.194 

0.995 

-0.75 % 

-0.36 % 

1.050 

0.399 

0.056 

0.995 

-0.33 % 

-1.62 % 

0.850 

0.399 

0.568 

0.995 

-0.33 % 

-0.89 % 

0.989 

0.399 

0.445 

0.995 

0.46 % 

0.68 % 

-0.502 

0.414 

-0.438 

0.995 

-0.22 % 

0.21 % 

0.620 

0.402 

-0.016 

0.995 

0.42 % 

1.2 % 

-0.563 

0.413 

-0.773 

0.995 

-0.5 % 

0.62 % 

0.802 

0.402 

-0.138 

0.995 

0.59 % 

0.61 % 

-1.127 

0.399 

-0.522 

0.995 

-0.07 % 

0.81 % 

0.227 

0.453 

-0.403 

0.995 

 286 

 
 
 
 
 
 
Table A.15 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_righ
t 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_righ
t 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

0.285 ± 
0.012 

0.283 ± 
0.010 

0.272 ± 
0.011 

0.285 ± 
0.011 

0.289 ± 
0.013 

0.295 ± 
0.013 

0.280 ± 
0.012 

0.296 ± 
0.013 

0.296 ± 
0.014 

0.311 ± 
0.016 

0.300 ± 
0.013 

0.304 ± 
0.013 

0.287 ± 
0.012 

0.277 ± 
0.011 

0.283 ± 
0.013 

0.286 ± 
0.014 

0.277 ± 
0.014 

0.275 ± 
0.013 

P2 

0.286 ± 
0.011 

0.283 ± 
0.011 

0.274 ± 
0.011 

0.285 ± 
0.013 

0.288 ± 
0.014 

0.294 ± 
0.014 

0.282 ± 
0.016 

0.296 ± 
0.016 

0.297 ± 
0.018 

0.310 ± 
0.023 

0.302 ± 
0.017 

0.305 ± 
0.018 

0.289 ± 
0.013 

0.278 ± 
0.013 

0.284 ± 
0.015 

0.289 ± 
0.016 

0.273 ± 
0.017 

0.271 ± 
0.018 

P1 

0.289 ± 
0.020 

0.284 ± 
0.014 

0.275 ± 
0.022 

0.285 ± 
0.021 

0.291 ± 
0.009 

0.294 ± 
0.014 

0.277 ± 
0.019 

0.292 ± 
0.020 

0.293 ± 
0.025 

0.305 ± 
0.033 

0.304 ± 
0.023 

0.302 ± 
0.024 

0.292 ± 
0.022 

0.278 ± 
0.017 

0.287 ± 
0.023 

0.288 ± 
0.027 

0.281 ± 
0.019 

0.277 ± 
0.014 

P2 

0.288 ± 
0.020 

0.286 ± 
0.019 

0.277 ± 
0.018 

0.285 ± 
0.024 

0.285 ± 
0.034 

0.291 ± 
0.025 

0.279 ± 
0.020 

0.294 ± 
0.020 

0.299 ± 
0.022 

0.307 ± 
0.029 

0.307 ± 
0.022 

0.305 ± 
0.020 

0.292 ± 
0.022 

0.282 ± 
0.016 

0.292 ± 
0.021 

0.291 ± 
0.023 

0.277 ± 
0.029 

0.276 ± 
0.033 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

0.38 % 

-0.36 % 

t-score 
mTBI 

-0.644 

q-value 
mTBI1 

0.402 

t-score 
controls 

q-value 
controls 

0.062 

0.995 

-0.05 % 

0.68 % 

0.382 

0.423 

-0.465 

0.995 

0.65 % 

0.5 % 

-1.035 

0.399 

-0.401 

0.995 

0.24 % 

0.16 % 

-0.431 

0.414 

-0.054 

0.995 

-0.4 % 

-1.79 % 

0.971 

0.399 

0.717 

0.995 

-0.18 % 

-0.98 % 

0.771 

0.402 

0.508 

0.995 

0.45 % 

0.71 % 

-0.459 

0.414 

-0.385 

0.995 

0.13 % 

0.5 % 

-0.039 

0.49 

-0.257 

0.995 

0.22 % 

2.07 % 

-0.162 

0.46 

-0.883 

0.995 

-0.27 % 

0.64 % 

0.288 

0.435 

-0.193 

0.995 

0.66 % 

1.05 % 

-1.180 

0.399 

-0.634 

0.995 

0.47 % 

0.96 % 

-0.665 

0.402 

-0.532 

0.995 

0.55 % 

0.11 % 

-0.909 

0.399 

-0.147 

0.995 

0.4 % 

1.23 % 

-0.668 

0.402 

-0.969 

0.995 

0.48 % 

1.63 % 

-0.571 

0.413 

-0.888 

0.995 

0.92 % 

1.01 % 

-1.308 

0.399 

-0.542 

0.995 

-1.6 % 

-1.17 % 

2.209 

0.399 

0.443 

0.995 

-1.45 % 

-0.37 % 

2.048 

0.399 

0.141 

0.995 

Tables A.15- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 287 

 
 
 
 
 
 
 
 
 
 
Table A.16 Results of Post Hoc Tract Specific Comparison of Diffusion Tensor Metric Mean 
Diffusivity   

mTBI 

Controls 

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

P1 

P2 

P1 

P2 

7.999e-04 ± 
2.322e-05 

7.868e-04 ± 
2.222e-05 

8.409e-04 ± 
3.346e-05 

8.461e-04 ± 
3.803e-05 

8.916e-04 ± 
2.541e-05 

8.586e-04 ± 
4.152e-05 

8.375e-04 ± 
2.624e-05 

8.569e-04 ± 
3.301e-05 

8.417e-04 ± 
2.701e-05 

8.335e-04 ± 
2.607e-05 

8.171e-04 ± 
3.205e-05 

8.307e-04 ± 
5.528e-05 

8.297e-04 ± 
2.690e-05 

7.949e-04 ± 
2.359e-05 

7.884e-04 ± 
2.595e-05 

8.246e-04 ± 
2.229e-05 

7.904e-04 ± 
2.320e-05 

8.283e-04 ± 
2.218e-05 

8.233e-04 ± 
2.398e-05 

1.609e-03 ± 
2.432e-04 

1.594e-03 ± 
2.327e-04 

7.459e-04 ± 
5.752e-05 

7.377e-04 ± 
4.462e-05 

8.088e-04 ± 
2.674e-05 

8.034e-04 ± 
3.017e-05 

8.232e-04 ± 
3.840e-05 

8.137e-04 ± 
3.955e-05 

7.943e-04 ± 
2.429e-05 

7.837e-04 ± 
1.881e-05 

8.358e-04 ± 
3.916e-05 

8.380e-04 ± 
3.574e-05 

8.887e-04 ± 
4.040e-05 

8.545e-04 ± 
4.578e-05 

8.373e-04 ± 
2.961e-05 

8.594e-04 ± 
4.634e-05 

8.466e-04 ± 
3.477e-05 

8.422e-04 ± 
3.891e-05 

8.205e-04 ± 
2.704e-05 

8.309e-04 ± 
3.870e-05 

8.312e-04 ± 
2.563e-05 

7.935e-04 ± 
2.323e-05 

7.859e-04 ± 
2.197e-05 

8.226e-04 ± 
3.099e-05 

7.882e-04 ± 
2.337e-05 

8.271e-04 ± 
3.052e-05 

8.223e-04 ± 
2.555e-05 

1.590e-03 ± 
2.420e-04 

1.579e-03 ± 
2.466e-04 

7.398e-04 ± 
3.820e-05 

7.329e-04 ± 
2.946e-05 

8.078e-04 ± 
2.595e-05 

8.009e-04 ± 
2.404e-05 

8.165e-04 ± 
3.108e-05 

8.085e-04 ± 
2.961e-05 

7.756e-04 ± 
3.105e-05 

7.764e-04 ± 
2.585e-05 

8.161e-04 ± 
4.401e-05 

8.330e-04 ± 
5.677e-05 

8.460e-04 ± 
3.763e-05 

8.494e-04 ± 
6.718e-05 

8.216e-04 ± 
3.730e-05 

8.383e-04 ± 
4.583e-05 

8.317e-04 ± 
4.394e-05 

8.266e-04 ± 
4.895e-05 

8.113e-04 ± 
3.214e-05 

8.254e-04 ± 
3.388e-05 

8.208e-04 ± 
3.323e-05 

7.818e-04 ± 
4.112e-05 

7.774e-04 ± 
4.034e-05 

7.999e-04 ± 
4.135e-05 

7.718e-04 ± 
3.429e-05 

8.068e-04 ± 
5.824e-05 

8.074e-04 ± 
5.361e-05 

1.597e-03 ± 
2.552e-04 

1.585e-03 ± 
2.478e-04 

7.409e-04 ± 
5.451e-05 

7.393e-04 ± 
5.818e-05 

7.904e-04 ± 
2.080e-05 

7.924e-04 ± 
2.364e-05 

8.006e-04 ± 
2.704e-05 

8.014e-04 ± 
3.538e-05 

7.686e-04 ± 
3.164e-05 

7.686e-04 ± 
2.719e-05 

8.005e-04 ± 
4.388e-05 

8.175e-04 ± 
5.208e-05 

8.296e-04 ± 
3.660e-05 

8.243e-04 ± 
4.685e-05 

8.096e-04 ± 
4.104e-05 

8.302e-04 ± 
6.241e-05 

8.367e-04 ± 
8.721e-05 

8.317e-04 ± 
9.654e-05 

8.161e-04 ± 
6.004e-05 

8.321e-04 ± 
6.310e-05 

8.188e-04 ± 
5.572e-05 

7.856e-04 ± 
7.225e-05 

7.734e-04 ± 
5.455e-05 

7.953e-04 ± 
6.590e-05 

7.795e-04 ± 
6.850e-05 

7.937e-04 ± 
6.377e-05 

7.924e-04 ± 
5.552e-05 

1.580e-03 ± 
2.864e-04 

1.529e-03 ± 
3.021e-04 

7.325e-04 ± 
5.920e-05 

7.150e-04 ± 
6.579e-05 

7.875e-04 ± 
3.704e-05 

7.865e-04 ± 
2.874e-05 

7.935e-04 ± 
4.082e-05 

7.931e-04 ± 
4.544e-05 

 288 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.7 % 

-0.91 % 

1.973 

0.54 

0.982 

0.601 

-0.4 % 

-1.01 % 

1.549 

0.54 

1.250 

0.596 

-0.61 % 

-1.91 % 

1.193 

0.54 

1.605 

0.596 

-0.95 % 

-1.86 % 

2.050 

0.54 

1.539 

0.596 

-0.33 % 

-1.95 % 

0.375 

0.795 

1.312 

0.596 

-0.48 % 

-2.96 % 

0.706 

0.722 

2.791 

0.44 

-0.02 % 

-1.45 % 

-0.114 

0.818 

1.151 

0.601 

0.29 % 

-0.97 % 

-0.480 

0.782 

0.590 

0.655 

0.58 % 

0.61 % 

-1.142 

0.548 

-0.062 

0.664 

1.04 % 

0.61 % 

-1.887 

0.54 

-0.106 

0.659 

0.41 % 

0.59 % 

-1.470 

0.54 

-0.202 

0.655 

0.02 % 

0.81 % 

-0.276 

0.804 

-0.200 

0.655 

0.19 % 

-0.25 % 

-0.877 

0.653 

0.323 

0.655 

-0.17 % 

0.49 % 

0.184 

0.818 

-0.162 

0.655 

-0.32 % 

-0.52 % 

0.506 

0.779 

0.346 

0.655 

-0.25 % 

-0.58 % 

0.540 

0.779 

0.391 

0.655 

-0.27 % 

0.99 % 

0.344 

0.795 

-0.414 

0.655 

-0.14 % 

-1.61 % 

0.082 

0.818 

0.754 

0.655 

-0.13 % 

-1.85 % 

0.064 

0.818 

1.147 

0.601 

-1.21 % 

-1.03 % 

1.273 

0.54 

0.502 

0.655 

-0.9 % 

-3.5 % 

0.615 

0.766 

1.423 

0.596 

-0.81 % 

-1.14 % 

0.751 

0.717 

0.436 

0.655 

-0.64 % 

-3.28 % 

0.681 

0.724 

1.396 

0.596 

-0.12 % 

-0.37 % 

0.146 

0.818 

0.541 

0.655 

-0.32 % 

-0.74 % 

0.516 

0.779 

2.517 

0.44 

-0.8 % 

-0.89 % 

1.882 

0.54 

1.041 

0.601 

-0.64 % 

-1.04 % 

1.299 

0.54 

2.062 

0.474 

 
 
 
 
 
 
 
 
Table A.16 (cont’d)   

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

7.251e-04 ± 
4.452e-05 

8.097e-04 ± 
2.705e-05 

8.024e-04 ± 
2.876e-05 

8.340e-04 ± 
3.934e-05 

8.257e-04 ± 
4.445e-05 

8.315e-04 ± 
2.473e-05 

8.162e-04 ± 
2.751e-05 

7.868e-04 ± 
2.581e-05 

7.779e-04 ± 
2.147e-05 

7.962e-04 ± 
2.382e-05 

7.828e-04 ± 
2.264e-05 

8.100e-04 ± 
2.939e-05 

8.013e-04 ± 
2.816e-05 

8.457e-04 ± 
3.854e-05 

8.284e-04 ± 
3.527e-05 

7.828e-04 ± 
2.238e-05 

7.658e-04 ± 
2.186e-05 

8.061e-04 ± 
3.489e-05 

8.106e-04 ± 
3.320e-05 

8.169e-04 ± 
3.222e-05 

8.034e-04 ± 
3.557e-05 

8.232e-04 ± 
2.826e-05 

8.055e-04 ± 
2.877e-05 

8.273e-04 ± 
2.845e-05 

7.981e-04 ± 
2.772e-05 

8.188e-04 ± 
2.658e-05 

7.950e-04 ± 
2.800e-05 

7.239e-04 ± 
3.381e-05 

8.088e-04 ± 
2.591e-05 

8.033e-04 ± 
2.571e-05 

8.315e-04 ± 
3.878e-05 

8.239e-04 ± 
3.876e-05 

8.350e-04 ± 
2.884e-05 

8.184e-04 ± 
2.682e-05 

7.823e-04 ± 
3.232e-05 

7.733e-04 ± 
2.205e-05 

7.944e-04 ± 
2.694e-05 

7.829e-04 ± 
2.113e-05 

8.057e-04 ± 
3.014e-05 

7.979e-04 ± 
2.601e-05 

8.478e-04 ± 
4.422e-05 

8.267e-04 ± 
3.849e-05 

7.788e-04 ± 
3.467e-05 

7.614e-04 ± 
2.653e-05 

8.052e-04 ± 
3.935e-05 

8.071e-04 ± 
3.231e-05 

8.147e-04 ± 
3.153e-05 

8.012e-04 ± 
3.058e-05 

8.248e-04 ± 
2.954e-05 

8.060e-04 ± 
2.623e-05 

8.256e-04 ± 
3.665e-05 

7.965e-04 ± 
2.909e-05 

8.139e-04 ± 
3.580e-05 

7.895e-04 ± 
2.563e-05 

7.178e-04 ± 
5.040e-05 

8.013e-04 ± 
3.068e-05 

7.993e-04 ± 
3.606e-05 

8.193e-04 ± 
3.161e-05 

8.261e-04 ± 
3.738e-05 

8.286e-04 ± 
4.189e-05 

8.094e-04 ± 
3.952e-05 

7.625e-04 ± 
4.451e-05 

7.682e-04 ± 
5.620e-05 

7.806e-04 ± 
5.033e-05 

7.793e-04 ± 
3.347e-05 

7.989e-04 ± 
4.336e-05 

7.935e-04 ± 
4.066e-05 

8.576e-04 ± 
8.340e-05 

8.296e-04 ± 
6.358e-05 

7.458e-04 ± 
4.054e-05 

7.499e-04 ± 
4.676e-05 

7.719e-04 ± 
3.447e-05 

7.906e-04 ± 
4.002e-05 

7.958e-04 ± 
1.883e-05 

7.937e-04 ± 
2.386e-05 

8.176e-04 ± 
3.713e-05 

7.978e-04 ± 
3.132e-05 

8.107e-04 ± 
4.887e-05 

7.923e-04 ± 
4.763e-05 

7.880e-04 ± 
4.243e-05 

7.793e-04 ± 
3.716e-05 

7.300e-04 ± 
6.087e-05 

7.968e-04 ± 
4.225e-05 

7.949e-04 ± 
4.326e-05 

8.179e-04 ± 
6.352e-05 

8.206e-04 ± 
5.220e-05 

8.256e-04 ± 
7.224e-05 

8.114e-04 ± 
6.759e-05 

7.543e-04 ± 
5.961e-05 

7.523e-04 ± 
3.878e-05 

7.732e-04 ± 
5.376e-05 

7.740e-04 ± 
4.087e-05 

7.891e-04 ± 
5.031e-05 

7.839e-04 ± 
4.224e-05 

8.492e-04 ± 
9.782e-05 

8.219e-04 ± 
8.344e-05 

7.504e-04 ± 
9.970e-05 

7.410e-04 ± 
6.010e-05 

7.685e-04 ± 
5.548e-05 

7.746e-04 ± 
2.859e-05 

7.951e-04 ± 
4.364e-05 

7.912e-04 ± 
3.384e-05 

8.152e-04 ± 
6.397e-05 

7.990e-04 ± 
5.421e-05 

8.032e-04 ± 
7.729e-05 

7.963e-04 ± 
7.827e-05 

7.863e-04 ± 
6.738e-05 

7.789e-04 ± 
5.685e-05 

 289 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-
score 
mTBI 

q-
value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.17 % 

1.7 % 

0.122  0.818 

-0.565 

0.655 

-0.1 % 

-0.56 % 

0.12 % 

-0.55 % 

-
0.066 

-
0.896 

0.818 

0.543 

0.655 

0.653 

0.539 

0.655 

-0.3 % 

-0.17 %  0.429  0.782 

0.143 

0.655 

-0.22 % 

-0.66 % 

0.41 % 

-0.36 % 

0.28 % 

0.24 % 

-
0.107 

-
1.469 

-
1.179 

0.818 

1.289 

0.596 

0.54 

0.275 

0.655 

0.54 

-0.131 

0.655 

-0.58 % 

-1.08 %  1.115  0.548 

0.432 

0.655 

-0.6 % 

-2.06 %  1.705 

0.54 

1.324 

0.596 

-0.23 % 

-0.95 %  0.702  0.722 

0.540 

0.655 

0.01 % 

-0.68 % 

-
0.056 

0.818 

0.518 

0.655 

-0.54 % 

-1.22 %  1.530 

0.54 

0.979 

0.601 

-0.43 % 

-1.22 %  1.100  0.548 

1.119 

0.601 

0.25 % 

-0.98 % 

-
0.461 

0.782 

0.456 

0.655 

-0.21 % 

-0.93 %  0.340  0.795 

0.525 

0.655 

-0.52 %  0.62 % 

0.994  0.617 

-0.254 

0.655 

-0.58 % 

-1.19 %  1.209 

0.54 

0.620 

0.655 

-0.11 % 

-0.44 % 

-
0.018 

0.823 

0.266 

0.655 

-0.44 % 

-2.03 %  0.979  0.617 

2.469 

0.44 

-0.27 % 

-0.1 % 

0.516  0.779 

0.222 

0.655 

-0.28 % 

-0.31 %  0.058  0.818 

2.300 

0.456 

0.19 % 

-0.3 % 

0.07 % 

0.15 % 

-
0.901 

-
0.747 

0.653 

0.333 

0.655 

0.717 

-0.003 

0.677 

-0.21 % 

-0.93 %  0.437  0.782 

0.529 

0.655 

-0.2 % 

0.5 % 

0.305  0.804 

-0.064 

0.664 

-0.6 % 

-0.22 %  1.414 

0.54 

0.261 

0.655 

-0.69 % 

-0.05 %  1.364 

0.54 

0.249 

0.655 

 
 
 
 
 
 
 
Table A.16 (cont’d)   

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

8.095e-04 ± 
2.509e-05 

8.088e-04 ± 
2.644e-05 

8.177e-04 ± 
2.850e-05 

8.057e-04 ± 
2.998e-05 

8.044e-04 ± 
2.507e-05 

8.119e-04 ± 
3.848e-05 

7.828e-04 ± 
4.072e-05 

7.893e-04 ± 
4.033e-05 

7.848e-04 ± 
4.277e-05 

7.759e-04 ± 
4.752e-05 

7.747e-04 ± 
3.372e-05 

7.662e-04 ± 
3.664e-05 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-
score 
mTBI 

q-
value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.47 % 

-0.88 %  1.338 

0.54 

0.623 

0.655 

-0.55 % 

-1.86 %  1.925 

0.54 

1.909 

0.537 

-0.71 % 

-2.38 %  1.669 

0.54 

2.073 

0.474 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

8.010e-04 ± 
2.579e-05 

8.390e-04 ± 
4.054e-05 

8.269e-04 ± 
4.470e-05 

8.334e-04 ± 
3.315e-05 

8.233e-04 ± 
3.552e-05 

8.257e-04 ± 
2.980e-05 

7.981e-04 ± 
3.203e-05 

8.208e-04 ± 
2.865e-05 

7.933e-04 ± 
3.078e-05 

8.255e-04 ± 
2.786e-05 

8.317e-04 ± 
3.290e-05 

8.370e-04 ± 
3.299e-05 

8.186e-04 ± 
3.557e-05 

8.116e-04 ± 
2.571e-05 

8.327e-04 ± 
2.444e-05 

7.946e-04 ± 
2.784e-05 

8.360e-04 ± 
3.969e-05 

8.247e-04 ± 
3.885e-05 

8.354e-04 ± 
3.495e-05 

8.233e-04 ± 
3.270e-05 

8.265e-04 ± 
3.850e-05 

7.993e-04 ± 
3.469e-05 

8.160e-04 ± 
3.953e-05 

7.874e-04 ± 
2.872e-05 

8.209e-04 ± 
3.525e-05 

8.253e-04 ± 
3.271e-05 

8.305e-04 ± 
4.623e-05 

8.100e-04 ± 
3.750e-05 

8.142e-04 ± 
2.716e-05 

8.318e-04 ± 
2.798e-05 

7.778e-04 ± 
4.761e-05 

8.243e-04 ± 
3.169e-05 

8.267e-04 ± 
3.792e-05 

8.344e-04 ± 
4.703e-05 

8.221e-04 ± 
4.550e-05 

8.250e-04 ± 
6.229e-05 

8.069e-04 ± 
6.816e-05 

7.968e-04 ± 
5.194e-05 

7.833e-04 ± 
4.619e-05 

8.027e-04 ± 
5.094e-05 

8.185e-04 ± 
5.609e-05 

8.121e-04 ± 
5.641e-05 

8.064e-04 ± 
7.571e-05 

7.926e-04 ± 
3.163e-05 

8.186e-04 ± 
4.089e-05 

7.692e-04 ± 
4.500e-05 

8.229e-04 ± 
6.250e-05 

8.218e-04 ± 
5.292e-05 

8.312e-04 ± 
7.984e-05 

8.210e-04 ± 
7.266e-05 

8.151e-04 ± 
9.048e-05 

8.105e-04 ± 
1.036e-04 

7.935e-04 ± 
8.025e-05 

7.813e-04 ± 
6.742e-05 

7.926e-04 ± 
5.755e-05 

8.013e-04 ± 
5.377e-05 

7.900e-04 ± 
6.052e-05 

7.926e-04 ± 
6.976e-05 

7.837e-04 ± 
2.922e-05 

8.018e-04 ± 
3.281e-05 

-0.8 % 

-1.11 % 

2.662 

0.54 

1.082 

0.601 

-0.35 % 

-0.17 % 

0.557 

0.779 

0.160 

0.655 

-0.26 % 

-0.6 % 

-0.005 

0.823 

1.278 

0.596 

0.24 % 

-0.39 % 

-0.767 

0.717 

0.304 

0.655 

0.01 % 

-0.13 % 

-0.345 

0.795 

0.170 

0.655 

0.09 % 

-1.2 % 

-0.128 

0.818 

0.548 

0.655 

0.15 % 

0.45 % 

-0.284 

0.804 

-0.007 

0.677 

-0.58 % 

-0.41 % 

1.258 

0.54 

0.296 

0.655 

-0.74 % 

-0.25 % 

1.218 

0.54 

0.296 

0.655 

-0.55 % 

-1.27 % 

1.190 

0.54 

0.732 

0.655 

-0.77 % 

-2.1 % 

1.805 

0.54 

1.514 

0.596 

-0.78 % 

-2.72 % 

1.289 

0.54 

1.551 

0.596 

-1.05 % 

-1.71 % 

1.890 

0.54 

1.009 

0.601 

0.33 % 

-1.12 % 

-1.175 

0.54 

0.985 

0.601 

-0.11 % 

-2.05 % 

0.048 

0.818 

1.435 

0.596 

Tables A.16- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 290 

 
 
 
 
 
 
 
 
 
 
 
 
Table A.17 Results of Post Hoc Tract Specific Comparison of Diffusion Tensor Metric Axial 
Diffusivity   

TractSeg Name 

P1 

P2 

P1 

P2 

mTBI 

Controls 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

1.015e-03 ± 
2.342e-05 

1.005e-03 ± 
2.231e-05 

1.079e-03 ± 
3.541e-05 

1.077e-03 ± 
4.010e-05 

1.153e-03 ± 
3.106e-05 

1.198e-03 ± 
5.031e-05 

1.109e-03 ± 
2.756e-05 

1.129e-03 ± 
3.908e-05 

1.121e-03 ± 
2.533e-05 

1.121e-03 ± 
2.864e-05 

1.083e-03 ± 
2.908e-05 

1.116e-03 ± 
5.499e-05 

1.097e-03 ± 
2.436e-05 

1.031e-03 ± 
2.378e-05 

1.027e-03 ± 
2.690e-05 

1.139e-03 ± 
2.442e-05 

1.100e-03 ± 
2.556e-05 

1.131e-03 ± 
2.590e-05 

1.124e-03 ± 
2.818e-05 

2.085e-03 ± 
2.114e-04 

2.077e-03 ± 
2.037e-04 

9.628e-04 ± 
5.867e-05 

9.386e-04 ± 
4.624e-05 

1.056e-03 ± 
2.808e-05 

1.047e-03 ± 
3.094e-05 

1.072e-03 ± 
4.333e-05 

1.053e-03 ± 
4.591e-05 

1.007e-03 ± 
2.573e-05 

1.000e-03 ± 
2.132e-05 

1.073e-03 ± 
4.054e-05 

1.068e-03 ± 
3.867e-05 

1.146e-03 ± 
3.747e-05 

1.187e-03 ± 
6.398e-05 

1.107e-03 ± 
3.109e-05 

1.129e-03 ± 
4.402e-05 

1.122e-03 ± 
3.542e-05 

1.126e-03 ± 
4.085e-05 

1.086e-03 ± 
3.002e-05 

1.114e-03 ± 
4.317e-05 

1.097e-03 ± 
2.547e-05 

1.032e-03 ± 
2.771e-05 

1.024e-03 ± 
2.496e-05 

1.136e-03 ± 
3.320e-05 

1.095e-03 ± 
2.623e-05 

1.129e-03 ± 
3.320e-05 

1.121e-03 ± 
2.925e-05 

2.046e-03 ± 
2.231e-04 

2.048e-03 ± 
2.239e-04 

9.534e-04 ± 
3.663e-05 

9.298e-04 ± 
3.252e-05 

1.053e-03 ± 
2.968e-05 

1.043e-03 ± 
2.640e-05 

1.061e-03 ± 
4.141e-05 

1.045e-03 ± 
3.940e-05 

9.847e-04 ± 
2.329e-05 

9.878e-04 ± 
2.429e-05 

1.053e-03 ± 
3.877e-05 

1.061e-03 ± 
5.729e-05 

1.113e-03 ± 
3.338e-05 

1.192e-03 ± 
7.785e-05 

1.087e-03 ± 
3.036e-05 

1.105e-03 ± 
3.968e-05 

1.097e-03 ± 
3.661e-05 

1.099e-03 ± 
4.358e-05 

1.068e-03 ± 
2.105e-05 

1.105e-03 ± 
4.256e-05 

1.080e-03 ± 
2.395e-05 

1.013e-03 ± 
3.399e-05 

1.009e-03 ± 
2.887e-05 

1.098e-03 ± 
3.481e-05 

1.068e-03 ± 
3.136e-05 

1.095e-03 ± 
4.753e-05 

1.095e-03 ± 
4.575e-05 

2.066e-03 ± 
2.208e-04 

2.052e-03 ± 
2.155e-04 

9.448e-04 ± 
4.185e-05 

9.265e-04 ± 
4.801e-05 

1.035e-03 ± 
2.153e-05 

1.033e-03 ± 
3.013e-05 

1.049e-03 ± 
3.176e-05 

1.043e-03 ± 
4.700e-05 

9.765e-04 ± 
2.670e-05 

9.759e-04 ± 
1.501e-05 

1.035e-03 ± 
3.740e-05 

1.044e-03 ± 
4.945e-05 

1.089e-03 ± 
2.505e-05 

1.152e-03 ± 
7.825e-05 

1.070e-03 ± 
2.602e-05 

1.092e-03 ± 
4.430e-05 

1.101e-03 ± 
6.714e-05 

1.103e-03 ± 
7.685e-05 

1.070e-03 ± 
4.447e-05 

1.104e-03 ± 
5.584e-05 

1.074e-03 ± 
3.965e-05 

1.013e-03 ± 
6.512e-05 

9.993e-04 ± 
4.918e-05 

1.092e-03 ± 
5.409e-05 

1.072e-03 ± 
5.302e-05 

1.078e-03 ± 
5.047e-05 

1.078e-03 ± 
4.495e-05 

2.028e-03 ± 
2.725e-04 

1.990e-03 ± 
3.089e-04 

9.368e-04 ± 
3.789e-05 

9.044e-04 ± 
7.191e-05 

1.024e-03 ± 
2.312e-05 

1.023e-03 ± 
3.073e-05 

1.030e-03 ± 
3.628e-05 

1.025e-03 ± 
6.098e-05 

 291 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.74 % 

-0.83 % 

2.557 

0.106 

1.569 

0.189 

-0.5 % 

-1.2 % 

2.287 

0.134 

2.777 

0.139 

-0.61 % 

-1.74 % 

1.545 

0.186 

2.169 

0.158 

-0.85 % 

-1.58 % 

2.237 

0.134 

1.727 

0.164 

-0.65 % 

-2.19 % 

1.287 

0.203 

2.472 

0.139 

-0.95 % 

-3.37 % 

1.590 

0.186 

1.825 

0.162 

-0.19 % 

-1.54 % 

0.586 

0.33 

2.309 

0.139 

-0.06 % 

-1.16 % 

0.102 

0.425 

1.115 

0.275 

0.16 % 

0.33 % 

-0.444 

0.363 

-0.064 

0.496 

0.47 % 

0.36 % 

-1.199 

0.219 

-0.134 

0.496 

0.25 % 

0.19 % 

-1.515 

0.188 

-0.054 

0.496 

-0.15 % 

-0.08 % 

0.192 

0.404 

0.985 

0.314 

0.02 % 

-0.58 % 

-0.389 

0.368 

1.020 

0.307 

0.11 % 

0.05 % 

-0.560 

0.33 

-0.038 

0.496 

-0.2 % 

-0.98 % 

0.231 

0.396 

0.789 

0.363 

-0.26 % 

-0.57 % 

0.809 

0.269 

0.682 

0.388 

-0.4 % 

0.41 % 

1.031 

0.219 

-0.248 

0.472 

-0.12 % 

-1.49 % 

0.280 

0.384 

1.297 

0.245 

-0.24 % 

-1.55 % 

0.644 

0.319 

1.833 

0.162 

-1.89 % 

-1.85 % 

2.755 

0.106 

0.802 

0.363 

-1.4 % 

-3.03 % 

1.849 

0.179 

1.194 

0.266 

-0.97 % 

-0.84 % 

1.186 

0.219 

0.644 

0.388 

-0.93 % 

-2.38 % 

1.472 

0.195 

1.293 

0.245 

-0.26 % 

-0.99 % 

0.835 

0.265 

3.486 

0.134 

-0.39 % 

-0.98 % 

1.302 

0.203 

2.324 

0.139 

-0.99 % 

-1.85 % 

2.310 

0.134 

2.306 

0.139 

-0.75 % 

-1.68 % 

1.675 

0.179 

1.883 

0.162 

 
 
 
 
 
 
 
 
Table A.17 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

9.671e-04 ± 
4.922e-05 

9.626e-04 ± 
3.853e-05 

9.479e-04 ± 
3.816e-05 

9.606e-04 ± 
4.000e-05 

1.033e-03 ± 
2.652e-05 

1.033e-03 ± 
2.832e-05 

1.020e-03 ± 
2.389e-05 

1.015e-03 ± 
3.585e-05 

1.029e-03 ± 
2.723e-05 

1.029e-03 ± 
2.783e-05 

1.019e-03 ± 
2.749e-05 

1.011e-03 ± 
3.368e-05 

1.102e-03 ± 
3.927e-05 

1.097e-03 ± 
4.030e-05 

1.084e-03 ± 
3.617e-05 

1.074e-03 ± 
4.706e-05 

1.094e-03 ± 
4.333e-05 

1.091e-03 ± 
3.987e-05 

1.093e-03 ± 
4.489e-05 

1.081e-03 ± 
6.077e-05 

1.109e-03 ± 
2.802e-05 

1.112e-03 ± 
3.210e-05 

1.094e-03 ± 
3.411e-05 

1.090e-03 ± 
6.348e-05 

1.100e-03 ± 
2.963e-05 

1.101e-03 ± 
3.086e-05 

1.084e-03 ± 
3.327e-05 

1.085e-03 ± 
5.636e-05 

1.057e-03 ± 
2.989e-05 

1.050e-03 ± 
3.242e-05 

1.016e-03 ± 
3.352e-05 

1.010e-03 ± 
3.981e-05 

1.044e-03 ± 
2.503e-05 

1.035e-03 ± 
2.552e-05 

1.018e-03 ± 
4.347e-05 

1.006e-03 ± 
2.895e-05 

1.013e-03 ± 
2.510e-05 

1.008e-03 ± 
2.923e-05 

9.897e-04 ± 
3.604e-05 

9.819e-04 ± 
4.455e-05 

1.006e-03 ± 
2.348e-05 

1.003e-03 ± 
2.646e-05 

9.898e-04 ± 
2.919e-05 

9.790e-04 ± 
2.266e-05 

1.017e-03 ± 
2.591e-05 

1.014e-03 ± 
2.883e-05 

1.001e-03 ± 
3.330e-05 

9.931e-04 ± 
4.407e-05 

1.007e-03 ± 
2.589e-05 

1.004e-03 ± 
2.526e-05 

9.925e-04 ± 
3.316e-05 

9.820e-04 ± 
3.410e-05 

1.051e-03 ± 
3.764e-05 

1.054e-03 ± 
4.073e-05 

1.056e-03 ± 
7.126e-05 

1.045e-03 ± 
8.891e-05 

1.041e-03 ± 
3.308e-05 

1.041e-03 ± 
3.473e-05 

1.036e-03 ± 
5.265e-05 

1.029e-03 ± 
7.461e-05 

1.071e-03 ± 
2.795e-05 

1.067e-03 ± 
3.761e-05 

1.018e-03 ± 
3.962e-05 

1.019e-03 ± 
8.800e-05 

1.052e-03 ± 
2.962e-05 

1.045e-03 ± 
3.229e-05 

1.025e-03 ± 
4.226e-05 

1.014e-03 ± 
5.503e-05 

1.059e-03 ± 
4.021e-05 

1.054e-03 ± 
4.817e-05 

1.025e-03 ± 
4.049e-05 

1.018e-03 ± 
3.978e-05 

1.073e-03 ± 
3.768e-05 

1.067e-03 ± 
3.943e-05 

1.051e-03 ± 
4.138e-05 

1.029e-03 ± 
2.899e-05 

1.082e-03 ± 
3.268e-05 

1.078e-03 ± 
3.508e-05 

1.055e-03 ± 
2.302e-05 

1.047e-03 ± 
2.724e-05 

ST_OCC_right 

1.062e-03 ± 
3.557e-05 

1.058e-03 ± 
3.316e-05 

1.049e-03 ± 
3.533e-05 

1.042e-03 ± 
3.978e-05 

ST_PAR_left 

ST_PAR_right 

1.064e-03 ± 
2.882e-05 

1.066e-03 ± 
3.255e-05 

1.050e-03 ± 
2.912e-05 

1.047e-03 ± 
5.636e-05 

1.055e-03 ± 
2.873e-05 

1.055e-03 ± 
2.928e-05 

1.040e-03 ± 
2.724e-05 

1.040e-03 ± 
4.490e-05 

ST_POSTC_left 

1.081e-03 ± 
3.349e-05 

1.078e-03 ± 
4.209e-05 

1.052e-03 ± 
4.335e-05 

1.043e-03 ± 
7.582e-05 

ST_POSTC_right  1.058e-03 ± 
3.114e-05 

1.053e-03 ± 
3.115e-05 

1.044e-03 ± 
4.490e-05 

1.046e-03 ± 
7.164e-05 

ST_PREC_left 

1.076e-03 ± 
2.947e-05 

1.070e-03 ± 
3.950e-05 

1.033e-03 ± 
3.791e-05 

1.029e-03 ± 
6.310e-05 

ST_PREC_right 

1.053e-03 ± 
3.045e-05 

1.046e-03 ± 
2.879e-05 

1.029e-03 ± 
3.861e-05 

1.026e-03 ± 
5.209e-05 

 292 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.47 % 

1.33 % 

0.633 

0.319 

-0.790 

0.363 

-0.05 % 

-0.51 % 

-0.337 

0.374 

0.790 

0.363 

0.01 % 

-0.8 % 

-0.683 

0.312 

1.284 

0.245 

-0.45 % 

-0.9 % 

1.057 

0.219 

1.332 

0.245 

-0.28 % 

-1.09 % 

0.313 

0.377 

1.722 

0.164 

0.28 % 

-0.34 % 

-1.430 

0.196 

0.428 

0.414 

0.11 % 

0.08 % 

-1.000 

0.219 

-0.040 

0.496 

-0.61 % 

-0.58 % 

1.567 

0.186 

0.556 

0.4 

-0.79 % 

-1.18 % 

3.230 

0.066 

1.782 

0.162 

-0.47 % 

-0.79 % 

1.828 

0.179 

0.710 

0.384 

-0.31 % 

-1.09 % 

1.030 

0.219 

2.007 

0.158 

-0.29 % 

-0.74 % 

1.177 

0.219 

0.873 

0.357 

-0.29 % 

-1.06 % 

1.045 

0.219 

1.652 

0.171 

0.32 % 

-1.01 % 

-1.142 

0.219 

0.632 

0.388 

-0.01 % 

-0.73 % 

-0.063 

0.427 

0.660 

0.388 

-0.42 % 

0.09 % 

1.108 

0.219 

-0.114 

0.496 

-0.64 % 

-1.03 % 

1.693 

0.179 

0.711 

0.384 

-0.47 % 

-0.7 % 

1.015 

0.219 

1.222 

0.262 

-0.6 % 

-2.05 % 

1.703 

0.179 

2.067 

0.158 

-0.36 % 

-0.73 % 

1.005 

0.219 

2.296 

0.139 

-0.32 % 

-0.69 % 

0.512 

0.342 

1.961 

0.161 

0.2 % 

-0.31 % 

-1.051 

0.219 

0.441 

0.414 

-0.05 % 

-0.05 % 

-0.383 

0.368 

0.190 

0.49 

-0.23 % 

-0.78 % 

0.560 

0.33 

0.565 

0.4 

-0.43 % 

0.22 % 

0.924 

0.239 

0.050 

0.496 

-0.51 % 

-0.4 % 

1.321 

0.203 

0.397 

0.421 

-0.72 % 

-0.23 % 

1.764 

0.179 

0.491 

0.414 

 
 
 
 
 
 
 
Table A.17 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_PREF_left 

1.056e-03 ± 
2.776e-05 

1.052e-03 ± 
3.464e-05 

1.022e-03 ± 
3.595e-05 

1.011e-03 ± 
4.110e-05 

ST_PREF_right 

1.054e-03 ± 
3.014e-05 

1.048e-03 ± 
3.014e-05 

1.028e-03 ± 
4.096e-05 

1.010e-03 ± 
3.036e-05 

ST_PREM_left 

1.053e-03 ± 
3.357e-05 

1.047e-03 ± 
4.465e-05 

1.010e-03 ± 
3.695e-05 

9.868e-04 ± 
2.854e-05 

ST_PREM_right 

1.044e-03 ± 
3.148e-05 

1.036e-03 ± 
3.357e-05 

1.011e-03 ± 
4.473e-05 

9.990e-04 ± 
4.066e-05 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

1.103e-03 ± 
4.013e-05 

1.097e-03 ± 
4.116e-05 

1.085e-03 ± 
3.587e-05 

1.076e-03 ± 
4.708e-05 

1.090e-03 ± 
4.353e-05 

1.086e-03 ± 
3.999e-05 

1.088e-03 ± 
4.596e-05 

1.077e-03 ± 
6.103e-05 

1.075e-03 ± 
3.197e-05 

1.078e-03 ± 
3.489e-05 

1.069e-03 ± 
3.948e-05 

1.064e-03 ± 
7.090e-05 

1.076e-03 ± 
3.398e-05 

1.076e-03 ± 
3.320e-05 

1.068e-03 ± 
3.920e-05 

1.066e-03 ± 
6.324e-05 

T_POSTC_left 

1.083e-03 ± 
3.168e-05 

1.083e-03 ± 
3.916e-05 

1.067e-03 ± 
5.405e-05 

1.058e-03 ± 
8.290e-05 

T_POSTC_right 

1.059e-03 ± 
3.156e-05 

1.059e-03 ± 
3.222e-05 

1.057e-03 ± 
5.480e-05 

1.059e-03 ± 
9.134e-05 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

1.083e-03 ± 
2.815e-05 

1.078e-03 ± 
3.880e-05 

1.047e-03 ± 
4.469e-05 

1.042e-03 ± 
7.164e-05 

1.049e-03 ± 
3.081e-05 

1.042e-03 ± 
2.871e-05 

1.031e-03 ± 
4.103e-05 

1.026e-03 ± 
6.015e-05 

1.076e-03 ± 
2.889e-05 

1.071e-03 ± 
3.634e-05 

1.047e-03 ± 
4.445e-05 

1.033e-03 ± 
4.993e-05 

1.074e-03 ± 
3.428e-05 

1.067e-03 ± 
3.426e-05 

1.055e-03 ± 
5.484e-05 

1.035e-03 ± 
4.852e-05 

1.087e-03 ± 
3.444e-05 

1.079e-03 ± 
4.639e-05 

1.054e-03 ± 
5.017e-05 

1.029e-03 ± 
5.125e-05 

T_PREM_right 

1.065e-03 ± 
3.573e-05 

1.057e-03 ± 
3.830e-05 

1.047e-03 ± 
6.767e-05 

1.030e-03 ± 
6.163e-05 

UF_left 

UF_right 

1.057e-03 ± 
2.840e-05 

1.056e-03 ± 
2.995e-05 

1.036e-03 ± 
3.224e-05 

1.020e-03 ± 
1.933e-05 

1.085e-03 ± 
2.861e-05 

1.080e-03 ± 
2.984e-05 

1.067e-03 ± 
4.207e-05 

1.045e-03 ± 
1.620e-05 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.4 % 

-1.1 % 

1.363 

0.196 

1.125 

0.275 

-0.55 % 

-1.72 % 

2.127 

0.15 

2.128 

0.158 

-0.57 % 

-2.27 % 

1.642 

0.179 

3.129 

0.136 

-0.71 % 

-1.21 % 

2.621 

0.106 

1.820 

0.162 

-0.48 % 

-0.9 % 

1.169 

0.219 

1.346 

0.245 

-0.3 % 

-1.03 % 

0.374 

0.368 

1.661 

0.171 

0.24 % 

-0.44 % 

-1.041 

0.219 

0.438 

0.414 

-0.02 % 

-0.25 % 

-0.368 

0.368 

0.311 

0.452 

-0.02 % 

-0.85 % 

0.156 

0.412 

0.599 

0.396 

-0.04 % 

0.14 % 

-0.040 

0.43 

0.071 

0.496 

-0.49 % 

-0.55 % 

1.405 

0.196 

0.462 

0.414 

-0.62 % 

-0.39 % 

1.424 

0.196 

0.501 

0.414 

-0.45 % 

-1.37 % 

1.370 

0.196 

1.152 

0.275 

-0.67 % 

-1.85 % 

2.012 

0.157 

1.799 

0.162 

-0.69 % 

-2.4 % 

1.718 

0.179 

2.040 

0.158 

-0.79 % 

-1.6 % 

2.079 

0.15 

1.369 

0.245 

-0.12 % 

-1.52 % 

0.079 

0.427 

2.707 

0.139 

-0.49 % 

-2.11 % 

1.658 

0.179 

2.610 

0.139 

Tables A.17- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 293 

 
 
 
 
 
 
 
 
 
 
 
Table A.18 Results of Post Hoc Tract Specific Comparison of Diffusion Tensor Metric Radial 
Diffusivity   

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

mTBI 

Controls 

P1 

P2 

P1 

P2 

6.926e-04 ± 
2.510e-05 

6.776e-04 ± 
2.346e-05 

7.217e-04 ± 
3.371e-05 

7.305e-04 ± 
3.776e-05 

7.607e-04 ± 
2.680e-05 

6.890e-04 ± 
4.217e-05 

7.017e-04 ± 
2.825e-05 

7.207e-04 ± 
3.389e-05 

7.022e-04 ± 
3.236e-05 

6.899e-04 ± 
3.043e-05 

6.840e-04 ± 
3.575e-05 

6.882e-04 ± 
5.660e-05 

6.962e-04 ± 
3.001e-05 

6.770e-04 ± 
2.722e-05 

6.694e-04 ± 
2.930e-05 

6.676e-04 ± 
2.487e-05 

6.358e-04 ± 
2.449e-05 

6.770e-04 ± 
2.380e-05 

6.729e-04 ± 
2.448e-05 

1.372e-03 ± 
2.615e-04 

1.352e-03 ± 
2.505e-04 

6.374e-04 ± 
5.814e-05 

6.372e-04 ± 
4.490e-05 

6.853e-04 ± 
2.739e-05 

6.818e-04 ± 
3.044e-05 

6.988e-04 ± 
3.708e-05 

6.943e-04 ± 
3.747e-05 

6.880e-04 ± 
2.491e-05 

6.754e-04 ± 
1.976e-05 

7.173e-04 ± 
3.941e-05 

7.230e-04 ± 
3.543e-05 

7.601e-04 ± 
4.472e-05 

6.885e-04 ± 
4.672e-05 

7.024e-04 ± 
3.106e-05 

7.247e-04 ± 
5.072e-05 

7.087e-04 ± 
3.842e-05 

7.003e-04 ± 
4.238e-05 

6.876e-04 ± 
2.890e-05 

6.893e-04 ± 
3.900e-05 

6.984e-04 ± 
2.789e-05 

6.744e-04 ± 
2.774e-05 

6.667e-04 ± 
2.393e-05 

6.660e-04 ± 
3.272e-05 

6.347e-04 ± 
2.548e-05 

6.760e-04 ± 
3.174e-05 

6.726e-04 ± 
2.664e-05 

1.362e-03 ± 
2.560e-04 

1.345e-03 ± 
2.628e-04 

6.330e-04 ± 
4.076e-05 

6.344e-04 ± 
2.976e-05 

6.851e-04 ± 
2.548e-05 

6.800e-04 ± 
2.425e-05 

6.941e-04 ± 
2.885e-05 

6.904e-04 ± 
2.834e-05 

6.710e-04 ± 
3.590e-05 

6.708e-04 ± 
2.842e-05 

6.977e-04 ± 
4.747e-05 

7.189e-04 ± 
5.714e-05 

7.125e-04 ± 
4.266e-05 

6.783e-04 ± 
6.571e-05 

6.889e-04 ± 
4.400e-05 

7.052e-04 ± 
5.606e-05 

6.990e-04 ± 
5.347e-05 

6.906e-04 ± 
5.817e-05 

6.831e-04 ± 
4.045e-05 

6.858e-04 ± 
3.193e-05 

6.912e-04 ± 
4.055e-05 

6.662e-04 ± 
4.741e-05 

6.615e-04 ± 
4.739e-05 

6.507e-04 ± 
4.765e-05 

6.237e-04 ± 
3.943e-05 

6.628e-04 ± 
6.547e-05 

6.634e-04 ± 
5.898e-05 

1.362e-03 ± 
2.757e-04 

1.351e-03 ± 
2.675e-04 

6.390e-04 ± 
6.150e-05 

6.457e-04 ± 
6.380e-05 

6.683e-04 ± 
2.269e-05 

6.720e-04 ± 
2.238e-05 

6.762e-04 ± 
2.638e-05 

6.808e-04 ± 
3.137e-05 

6.646e-04 ± 
3.528e-05 

6.650e-04 ± 
3.511e-05 

6.835e-04 ± 
4.824e-05 

7.041e-04 ± 
5.440e-05 

7.000e-04 ± 
4.786e-05 

6.606e-04 ± 
5.164e-05 

6.794e-04 ± 
5.003e-05 

6.995e-04 ± 
7.412e-05 

7.047e-04 ± 
9.832e-05 

6.961e-04 ± 
1.081e-04 

6.892e-04 ± 
6.888e-05 

6.963e-04 ± 
7.274e-05 

6.913e-04 ± 
6.451e-05 

6.717e-04 ± 
7.710e-05 

6.604e-04 ± 
5.967e-05 

6.469e-04 ± 
7.265e-05 

6.331e-04 ± 
7.721e-05 

6.514e-04 ± 
7.116e-05 

6.495e-04 ± 
6.193e-05 

1.356e-03 ± 
3.019e-04 

1.299e-03 ± 
3.085e-04 

6.304e-04 ± 
7.152e-05 

6.203e-04 ± 
6.577e-05 

6.691e-04 ± 
4.612e-05 

6.683e-04 ± 
3.391e-05 

6.752e-04 ± 
4.890e-05 

6.771e-04 ± 
4.371e-05 

 294 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.66 % 

-0.96 % 

1.503 

0.963 

0.792 

0.987 

-0.32 % 

-0.87 % 

0.827 

0.963 

0.752 

0.987 

-0.61 % 

-2.03 % 

0.991 

0.963 

1.354 

0.987 

-1.02 % 

-2.06 % 

1.871 

0.963 

1.417 

0.987 

-0.08 % 

-1.75 % 

-0.072 

0.987 

0.856 

0.987 

-0.07 % 

-2.6 % 

-0.144 

0.987 

1.762 

0.987 

0.11 % 

-1.39 % 

-0.448 

0.987 

0.806 

0.987 

0.55 % 

-0.81 % 

-0.687 

0.987 

0.418 

0.987 

0.93 % 

0.82 % 

-1.303 

0.963 

-0.061 

0.987 

1.51 % 

0.81 % 

-1.989 

0.963 

-0.094 

0.987 

0.53 % 

0.89 % 

-1.273 

0.963 

-0.237 

0.987 

0.16 % 

1.53 % 

-0.473 

0.987 

-0.442 

0.987 

0.31 % 

0.01 % 

-0.949 

0.963 

0.150 

0.987 

-0.4 % 

0.83 % 

0.451 

0.987 

-0.211 

0.987 

-0.4 % 

-0.17 % 

0.534 

0.987 

0.194 

0.987 

-0.24 % 

-0.6 % 

0.375 

0.987 

0.288 

0.987 

-0.17 % 

1.5 % 

0.037 

0.987 

-0.458 

0.987 

-0.15 % 

-1.72 % 

-0.016 

0.987 

0.573 

0.987 

-0.04 % 

-2.1 % 

-0.186 

0.987 

0.911 

0.987 

-0.7 % 

-0.4 % 

0.530 

0.987 

0.313 

0.987 

-0.51 % 

-3.85 % 

0.056 

0.987 

1.528 

0.987 

-0.69 % 

-1.35 % 

0.480 

0.987 

0.371 

0.987 

-0.43 % 

-3.93 % 

0.206 

0.987 

1.382 

0.987 

-0.02 % 

0.11 % 

-0.248 

0.987 

0.016 

0.987 

-0.26 % 

-0.54 % 

0.072 

0.987 

1.036 

0.987 

-0.66 % 

-0.15 % 

1.299 

0.963 

0.221 

0.987 

-0.56 % 

-0.55 % 

0.857 

0.963 

1.387 

0.987 

 
 
 
 
 
 
 
 
Table A.18 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

6.041e-04 ± 
4.311e-05 

6.978e-04 ± 
2.931e-05 

6.892e-04 ± 
3.101e-05 

7.003e-04 ± 
4.076e-05 

6.915e-04 ± 
4.588e-05 

6.928e-04 ± 
2.568e-05 

6.742e-04 ± 
2.835e-05 

6.519e-04 ± 
2.732e-05 

6.451e-04 ± 
2.198e-05 

6.876e-04 ± 
2.547e-05 

6.713e-04 ± 
2.392e-05 

7.067e-04 ± 
3.277e-05 

6.984e-04 ± 
3.077e-05 

7.430e-04 ± 
4.054e-05 

7.220e-04 ± 
3.788e-05 

6.387e-04 ± 
2.448e-05 

6.227e-04 ± 
2.252e-05 

6.794e-04 ± 
3.460e-05 

6.794e-04 ± 
3.338e-05 

6.844e-04 ± 
3.375e-05 

6.743e-04 ± 
3.673e-05 

7.027e-04 ± 
2.979e-05 

6.805e-04 ± 
3.013e-05 

7.006e-04 ± 
2.844e-05 

6.683e-04 ± 
2.854e-05 

6.905e-04 ± 
2.735e-05 

6.658e-04 ± 
2.834e-05 

6.046e-04 ± 
3.286e-05 

6.028e-04 ± 
5.747e-05 

6.147e-04 ± 
7.219e-05 

6.968e-04 ± 
2.710e-05 

6.920e-04 ± 
3.567e-05 

6.879e-04 ± 
4.632e-05 

6.905e-04 ± 
2.741e-05 

6.895e-04 ± 
4.188e-05 

6.870e-04 ± 
4.959e-05 

6.990e-04 ± 
3.968e-05 

6.871e-04 ± 
3.076e-05 

6.899e-04 ± 
7.430e-05 

6.903e-04 ± 
3.964e-05 

6.925e-04 ± 
3.584e-05 

6.903e-04 ± 
5.206e-05 

6.964e-04 ± 
3.043e-05 

6.958e-04 ± 
4.768e-05 

6.931e-04 ± 
7.727e-05 

6.769e-04 ± 
2.844e-05 

6.720e-04 ± 
4.540e-05 

6.745e-04 ± 
7.402e-05 

6.483e-04 ± 
3.494e-05 

6.356e-04 ± 
5.143e-05 

6.262e-04 ± 
7.020e-05 

6.422e-04 ± 
2.353e-05 

6.433e-04 ± 
6.362e-05 

6.256e-04 ± 
4.642e-05 

6.873e-04 ± 
2.821e-05 

6.761e-04 ± 
5.829e-05 

6.689e-04 ± 
5.911e-05 

6.730e-04 ± 
2.385e-05 

6.741e-04 ± 
3.759e-05 

6.716e-04 ± 
5.256e-05 

7.016e-04 ± 
3.297e-05 

6.981e-04 ± 
4.919e-05 

6.871e-04 ± 
5.678e-05 

6.947e-04 ± 
2.901e-05 

6.940e-04 ± 
4.574e-05 

6.848e-04 ± 
4.679e-05 

7.445e-04 ± 
4.786e-05 

7.586e-04 ± 
9.004e-05 

7.513e-04 ± 
1.027e-04 

7.195e-04 ± 
4.255e-05 

7.262e-04 ± 
7.028e-05 

7.184e-04 ± 
8.837e-05 

6.349e-04 ± 
3.693e-05 

6.095e-04 ± 
4.429e-05 

6.160e-04 ± 
1.065e-04 

6.193e-04 ± 
2.865e-05 

6.126e-04 ± 
5.202e-05 

6.044e-04 ± 
6.454e-05 

6.807e-04 ± 
3.755e-05 

6.454e-04 ± 
3.381e-05 

6.439e-04 ± 
6.515e-05 

6.773e-04 ± 
3.279e-05 

6.604e-04 ± 
4.087e-05 

6.472e-04 ± 
3.854e-05 

6.831e-04 ± 
3.157e-05 

6.662e-04 ± 
2.020e-05 

6.689e-04 ± 
5.584e-05 

6.726e-04 ± 
3.110e-05 

6.658e-04 ± 
2.164e-05 

6.658e-04 ± 
3.611e-05 

7.040e-04 ± 
3.110e-05 

7.013e-04 ± 
4.253e-05 

6.992e-04 ± 
6.829e-05 

6.816e-04 ± 
2.781e-05 

6.765e-04 ± 
3.588e-05 

6.785e-04 ± 
5.953e-05 

6.993e-04 ± 
3.708e-05 

6.903e-04 ± 
5.255e-05 

6.831e-04 ± 
7.874e-05 

6.682e-04 ± 
3.251e-05 

6.665e-04 ± 
5.109e-05 

6.713e-04 ± 
8.250e-05 

6.859e-04 ± 
3.627e-05 

6.655e-04 ± 
4.600e-05 

6.650e-04 ± 
7.016e-05 

6.613e-04 ± 
2.734e-05 

6.546e-04 ± 
3.867e-05 

6.551e-04 ± 
6.024e-05 

 295 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

0.08 % 

1.98 % 

-0.163 

0.987 

-0.491 

0.987 

-0.14 % 

-0.6 % 

0.069 

0.987 

0.457 

0.987 

0.19 % 

-0.36 % 

-0.874 

0.963 

0.337 

0.987 

-0.19 % 

0.41 % 

0.115 

0.987 

-0.177 

0.987 

-0.18 % 

-0.33 % 

-0.281 

0.987 

0.761 

0.987 

0.52 % 

-0.38 % 

-1.306 

0.963 

0.219 

0.987 

0.41 % 

0.37 % 

-1.153 

0.963 

-0.159 

0.987 

-0.55 % 

-1.47 % 

0.747 

0.987 

0.392 

0.987 

-0.44 % 

-2.76 % 

0.798 

0.963 

1.152 

0.987 

-0.05 % 

-1.07 % 

0.140 

0.987 

0.484 

0.987 

0.25 % 

-0.38 % 

-0.618 

0.987 

0.204 

0.987 

-0.72 % 

-1.56 % 

1.508 

0.963 

0.954 

0.987 

-0.53 % 

-1.33 % 

0.963 

0.963 

0.941 

0.987 

0.19 % 

-0.96 % 

-0.223 

0.987 

0.386 

0.987 

-0.35 % 

-1.07 % 

0.451 

0.987 

0.476 

0.987 

-0.6 % 

1.07 % 

0.837 

0.963 

-0.314 

0.987 

-0.53 % 

-1.33 % 

0.894 

0.963 

0.571 

0.987 

0.18 % 

-0.23 % 

-0.706 

0.987 

0.027 

0.987 

-0.31 % 

-2.01 % 

0.394 

0.987 

1.637 

0.987 

-0.2 % 

0.41 % 

0.179 

0.987 

-0.164 

0.987 

-0.25 % 

-0.01 % 

-0.162 

0.987 

0.772 

0.987 

0.18 % 

-0.3 % 

-0.660 

0.987 

0.291 

0.987 

0.16 % 

0.3 % 

-0.819 

0.963 

-0.066 

0.987 

-0.19 % 

-1.05 % 

0.317 

0.987 

0.506 

0.987 

-0.03 % 

0.72 % 

-0.032 

0.987 

-0.110 

0.987 

-0.66 % 

-0.07 % 

1.339 

0.963 

0.198 

0.987 

-0.68 % 

0.08 % 

1.044 

0.963 

0.145 

0.987 

 
 
 
 
 
 
 
Table A.18 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

6.862e-04 ± 
2.550e-05 

6.861e-04 ± 
2.578e-05 

7.003e-04 ± 
2.757e-05 

6.798e-04 ± 
2.468e-05 

7.072e-04 ± 
4.198e-05 

6.955e-04 ± 
4.611e-05 

7.125e-04 ± 
3.498e-05 

6.970e-04 ± 
3.743e-05 

6.971e-04 ± 
3.117e-05 

6.677e-04 ± 
3.483e-05 

6.898e-04 ± 
3.095e-05 

6.655e-04 ± 
3.227e-05 

7.001e-04 ± 
2.893e-05 

7.107e-04 ± 
3.313e-05 

7.121e-04 ± 
3.405e-05 

6.954e-04 ± 
3.690e-05 

6.887e-04 ± 
2.637e-05 

6.826e-04 ± 
2.901e-05 

6.631e-04 ± 
4.479e-05 

6.584e-04 ± 
5.170e-05 

6.823e-04 ± 
2.420e-05 

6.701e-04 ± 
4.125e-05 

6.570e-04 ± 
3.744e-05 

6.946e-04 ± 
3.653e-05 

6.724e-04 ± 
4.717e-05 

6.559e-04 ± 
4.144e-05 

6.738e-04 ± 
2.700e-05 

6.611e-04 ± 
5.068e-05 

6.543e-04 ± 
4.936e-05 

7.054e-04 ± 
4.044e-05 

6.937e-04 ± 
3.080e-05 

6.965e-04 ± 
7.268e-05 

6.939e-04 ± 
3.958e-05 

6.959e-04 ± 
3.580e-05 

6.940e-04 ± 
5.270e-05 

7.142e-04 ± 
3.744e-05 

7.171e-04 ± 
5.163e-05 

7.146e-04 ± 
8.467e-05 

6.972e-04 ± 
3.505e-05 

6.990e-04 ± 
4.984e-05 

6.988e-04 ± 
7.777e-05 

6.983e-04 ± 
4.104e-05 

7.039e-04 ± 
6.754e-05 

6.936e-04 ± 
9.480e-05 

6.697e-04 ± 
3.996e-05 

6.816e-04 ± 
7.665e-05 

6.863e-04 ± 
1.103e-04 

6.852e-04 ± 
4.194e-05 

6.714e-04 ± 
5.676e-05 

6.694e-04 ± 
8.506e-05 

6.598e-04 ± 
3.168e-05 

6.596e-04 ± 
5.046e-05 

6.588e-04 ± 
7.162e-05 

6.957e-04 ± 
3.602e-05 

6.806e-04 ± 
5.524e-05 

6.725e-04 ± 
6.216e-05 

7.047e-04 ± 
3.333e-05 

7.004e-04 ± 
5.745e-05 

6.844e-04 ± 
5.691e-05 

7.060e-04 ± 
4.735e-05 

6.912e-04 ± 
6.029e-05 

6.707e-04 ± 
6.564e-05 

6.867e-04 ± 
3.860e-05 

6.861e-04 ± 
8.031e-05 

6.738e-04 ± 
7.435e-05 

6.933e-04 ± 
2.901e-05 

6.710e-04 ± 
3.417e-05 

6.656e-04 ± 
3.905e-05 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.53 % 

-0.71 % 

1.215 

0.963 

0.428 

0.987 

-0.55 % 

-1.96 % 

1.547 

0.963 

1.664 

0.987 

-0.81 % 

-2.45 % 

1.588 

0.963 

1.643 

0.987 

-0.88 % 

-1.02 % 

2.357 

0.963 

0.792 

0.987 

-0.25 % 

0.4 % 

0.243 

0.987 

-0.168 

0.987 

-0.23 % 

-0.27 % 

-0.166 

0.987 

0.784 

0.987 

0.24 % 

-0.35 % 

-0.595 

0.987 

0.251 

0.987 

0.03 % 

-0.04 % 

-0.318 

0.987 

0.118 

0.987 

0.18 % 

-1.46 % 

-0.242 

0.987 

0.524 

0.987 

0.3 % 

0.68 % 

-0.364 

0.987 

-0.037 

0.987 

-0.66 % 

-0.31 % 

1.137 

0.963 

0.225 

0.987 

-0.84 % 

-0.13 % 

1.085 

0.963 

0.213 

0.987 

-0.63 % 

-1.19 % 

1.064 

0.963 

0.560 

0.987 

-0.85 % 

-2.29 % 

1.625 

0.963 

1.371 

0.987 

-0.85 % 

-2.97 % 

1.082 

0.963 

1.337 

0.987 

-1.25 % 

-1.8 % 

1.770 

0.963 

0.869 

0.987 

0.67 % 

-0.81 % 

-1.624 

0.963 

0.422 

0.987 

UF_right 

7.077e-04 ± 
3.061e-05 

7.064e-04 ± 
2.464e-05 

6.804e-04 ± 
4.518e-05 
Tables A.18- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

6.943e-04 ± 
4.186e-05 

-2.01 % 

0.19 % 

-0.641 

0.987 

0.938 

0.987 

 296 

 
 
 
 
 
 
 
 
 
 
 
Table A.19 Results of Post Hoc Tract Specific Comparison of Diffusion Kurtosis Imaging 
Kurtosis Fractional Anisotropy 

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls % 

t-
score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

0.200 ± 0.010 

0.200 ± 0.008 

0.202 ± 0.014 

0.204 ± 0.013 

0.02 % 

0.77 % 

0.046 

0.456 

-0.557 

0.204 ± 0.008 

0.203 ± 0.008 

0.202 ± 0.011 

0.201 ± 0.020 

-0.53 % 

-0.71 % 

1.520 

0.422 

0.213 

0.204 ± 0.008 

0.204 ± 0.009 

0.210 ± 0.015 

0.213 ± 0.014 

0.26 % 

1.53 % 

-0.365 

0.422 

-0.987 

0.199 ± 0.008 

0.200 ± 0.009 

0.203 ± 0.013 

0.204 ± 0.016 

0.27 % 

0.77 % 

-0.594 

0.422 

-0.503 

0.208 ± 0.013 

0.207 ± 0.018 

0.220 ± 0.016 

0.219 ± 0.027 

-0.34 % 

-0.44 % 

0.470 

0.422 

0.258 ± 0.016 

0.255 ± 0.023 

0.264 ± 0.017 

0.257 ± 0.041 

-0.98 % 

-2.65 % 

1.215 

0.422 

0.014 

0.517 

0.224 ± 0.010 

0.223 ± 0.010 

0.227 ± 0.016 

0.228 ± 0.021 

-0.34 % 

0.43 % 

0.874 

0.422 

-0.350 

0.224 ± 0.012 

0.223 ± 0.017 

0.229 ± 0.025 

0.231 ± 0.028 

-0.67 % 

0.234 ± 0.013 

0.232 ± 0.014 

0.235 ± 0.024 

0.238 ± 0.025 

-0.64 % 

1.07 % 

1.09 % 

0.759 

0.422 

-0.464 

0.921 

0.422 

-0.431 

0.243 ± 0.013 

0.241 ± 0.015 

0.243 ± 0.025 

0.245 ± 0.029 

-0.98 % 

1.06 % 

1.382 

0.422 

-0.362 

0.234 ± 0.013 

0.234 ± 0.010 

0.230 ± 0.017 

0.229 ± 0.021 

-0.27 % 

-0.14 % 

0.731 

0.422 

-0.014 

0.236 ± 0.012 

0.235 ± 0.012 

0.233 ± 0.009 

0.229 ± 0.030 

-0.33 % 

-1.92 % 

0.939 

0.422 

0.555 

0.228 ± 0.010 

0.228 ± 0.009 

0.228 ± 0.016 

0.228 ± 0.021 

-0.3 % 

0.06 % 

0.859 

0.422 

-0.142 

0.215 ± 0.012 

0.217 ± 0.014 

0.217 ± 0.018 

0.216 ± 0.022 

0.95 % 

-0.37 % 

-0.737 

0.422 

-0.025 

0.219 ± 0.013 

0.219 ± 0.012 

0.218 ± 0.017 

0.217 ± 0.019 

0.21 % 

-0.16 % 

-0.217 

0.422 

-0.218 

0.255 ± 0.010 

0.257 ± 0.011 

0.260 ± 0.019 

0.264 ± 0.018 

0.45 % 

1.47 % 

-0.880 

0.422 

-0.717 

0.262 ± 0.010 

0.262 ± 0.011 

0.263 ± 0.018 

0.263 ± 0.024 

0.07 % 

0.09 % 

-0.048 

0.456 

-0.030 

0.246 ± 0.010 

0.247 ± 0.011 

0.249 ± 0.020 

0.252 ± 0.020 

0.44 % 

1.2 % 

-0.573 

0.422 

-0.558 

0.248 ± 0.010 

0.248 ± 0.010 

0.249 ± 0.017 

0.251 ± 0.020 

0.15 % 

1 % 

-0.234 

0.422 

-0.473 

0.185 ± 0.037 

0.186 ± 0.039 

0.192 ± 0.038 

0.191 ± 0.046 

0.55 % 

-0.54 % 

-0.221 

0.422 

-0.212 

0.189 ± 0.037 

0.190 ± 0.042 

0.193 ± 0.039 

0.198 ± 0.042 

0.56 % 

2.1 % 

-0.065 

0.456 

-1.034 

0.200 ± 0.020 

0.200 ± 0.022 

0.189 ± 0.023 

0.189 ± 0.031 

-0.15 % 

-0.38 % 

0.463 

0.422 

0.090 

0.189 ± 0.013 

0.187 ± 0.014 

0.178 ± 0.019 

0.181 ± 0.024 

-1.15 % 

1.79 % 

1.833 

0.422 

-0.365 

0.217 ± 0.009 

0.216 ± 0.009 

0.218 ± 0.010 

0.215 ± 0.023 

-0.26 % 

-1.35 % 

0.930 

0.422 

0.216 ± 0.008 

0.215 ± 0.009 

0.216 ± 0.008 

0.215 ± 0.021 

-0.37 % 

-0.7 % 

1.343 

0.422 

0.214 ± 0.009 

0.213 ± 0.011 

0.217 ± 0.007 

0.212 ± 0.027 

-0.45 % 

-2.37 % 

0.807 

0.422 

0.214 ± 0.009 

0.213 ± 0.013 

0.216 ± 0.009 

0.212 ± 0.028 

-0.29 % 

-1.6 % 

0.839 

0.422 

0.218 ± 0.012 

0.217 ± 0.012 

0.212 ± 0.023 

0.209 ± 0.026 

-0.53 % 

-1.08 % 

1.023 

0.422 

0.418 

0.358 

0.742 

0.489 

0.280 

0.207 ± 0.011 

0.208 ± 0.008 

0.204 ± 0.014 

0.206 ± 0.013 

0.33 % 

0.85 % 

-0.362 

0.422 

-0.552 

0.214 ± 0.010 

0.213 ± 0.009 

0.209 ± 0.015 

0.208 ± 0.019 

-0.46 % 

-0.05 % 

1.247 

0.422 

-0.061 

0.223 ± 0.011 

0.223 ± 0.011 

0.223 ± 0.007 

0.220 ± 0.028 

-0.08 % 

-1.31 % 

0.308 

0.422 

0.231 ± 0.011 

0.231 ± 0.011 

0.230 ± 0.009 

0.228 ± 0.020 

-0.09 % 

-0.66 % 

0.644 

0.422 

0.474 

0.321 

0.238 ± 0.009 

0.238 ± 0.010 

0.236 ± 0.017 

0.239 ± 0.017 

0.12 % 

1.2 % 

-0.027 

0.456 

-0.549 

0.249 ± 0.010 

0.249 ± 0.010 

0.248 ± 0.017 

0.250 ± 0.020 

-0.06 % 

0.78 % 

0.469 

0.422 

-0.236 

0.230 ± 0.013 

0.231 ± 0.017 

0.229 ± 0.019 

0.233 ± 0.023 

0.4 % 

1.81 % 

-0.393 

0.422 

-0.629 

0.232 ± 0.009 

0.231 ± 0.012 

0.226 ± 0.020 

0.231 ± 0.021 

-0.32 % 

2.19 % 

0.672 

0.422 

-0.618 

SLF_III_left 

0.205 ± 0.011 

0.204 ± 0.010 

0.205 ± 0.021 

0.208 ± 0.017 

-0.62 % 

1.43 % 

0.985 

0.422 

-0.661 

0.629 

0.629 

0.629 

0.629 

0.662 

0.629 

0.629 

0.629 

0.629 

0.629 

0.662 

0.629 

0.649 

0.662 

0.629 

0.629 

0.662 

0.629 

0.629 

0.629 

0.629 

0.662 

0.629 

0.629 

0.629 

0.629 

0.629 

0.629 

0.629 

0.662 

0.629 

0.629 

0.629 

0.629 

0.629 

0.629 

0.629 

 297 

 
 
 
 
 
 
 
 
Table A.19 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-
score 
mTBI 

q-
value 
mTBI1 

t-score 
controls 

q-
value 
contro
ls 

SLF_III_right 

0.209 ± 0.009  0.206 ± 0.011  0.203 ± 0.014 

0.200 ± 0.028 

-1.33 % 

-1.53 % 

2.162 

0.422 

0.388 

0.629 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

0.193 ± 0.011  0.194 ± 0.011  0.192 ± 0.015 

0.198 ± 0.018 

0.69 % 

2.71 % 

-0.807  0.422 

-1.156 

0.629 

0.194 ± 0.010  0.194 ± 0.010  0.191 ± 0.014 

0.194 ± 0.015 

0.08 % 

1.46 % 

0.322 

0.422 

-0.755 

0.629 

0.193 ± 0.010  0.194 ± 0.012  0.190 ± 0.022 

0.193 ± 0.018 

0.44 % 

1.79 % 

-0.511  0.422 

-0.686 

0.629 

0.201 ± 0.011  0.203 ± 0.012  0.199 ± 0.019 

0.203 ± 0.017 

0.58 % 

2.23 % 

-0.511  0.422 

-0.775 

0.629 

0.247 ± 0.011  0.249 ± 0.014  0.254 ± 0.019 

0.256 ± 0.026 

0.53 % 

1.05 % 

-0.822  0.422 

-0.477 

0.629 

0.255 ± 0.012  0.255 ± 0.012  0.257 ± 0.020 

0.261 ± 0.018 

0.01 % 

1.7 % 

-0.269  0.422 

-0.996 

0.629 

0.225 ± 0.012  0.223 ± 0.013  0.233 ± 0.013 

0.231 ± 0.026 

-0.77 % 

-0.72 % 

1.334 

0.422 

0.228 ± 0.013  0.227 ± 0.016  0.232 ± 0.014 

0.229 ± 0.028 

-0.52 % 

-1.02 % 

0.891 

0.422 

0.225 ± 0.011  0.225 ± 0.011  0.225 ± 0.010 

0.221 ± 0.028 

0 % 

-1.4 % 

0.250 

0.422 

0.355 

0.191 

0.447 

0.250 

0.629 

0.631 

0.629 

0.629 

ST_OCC_right 

0.229 ± 0.011  0.228 ± 0.011  0.227 ± 0.010 

0.226 ± 0.020 

-0.34 % 

-0.46 % 

1.063 

0.422 

ST_PAR_left 

0.217 ± 0.010  0.218 ± 0.009  0.215 ± 0.016 

0.217 ± 0.015 

0.42 % 

1.09 % 

-0.484  0.422 

-0.645 

0.629 

ST_PAR_right 

0.228 ± 0.010  0.228 ± 0.009  0.226 ± 0.014 

0.228 ± 0.016 

-0.25 % 

0.76 % 

1.006 

0.422 

-0.265 

0.629 

ST_POSTC_left 

0.222 ± 0.010  0.223 ± 0.011  0.226 ± 0.017 

0.230 ± 0.015 

0.57 % 

1.82 % 

-0.816  0.422 

-0.983 

0.629 

ST_POSTC_right 

0.234 ± 0.011  0.233 ± 0.012  0.235 ± 0.017 

0.238 ± 0.019 

-0.22 % 

1.24 % 

0.606 

0.422 

-0.434 

0.629 

ST_PREC_left 

0.222 ± 0.009  0.223 ± 0.011  0.229 ± 0.016 

0.231 ± 0.014 

0.53 % 

1.18 % 

-1.044  0.422 

-0.802 

0.629 

ST_PREC_right 

0.230 ± 0.009  0.230 ± 0.009  0.231 ± 0.015 

0.234 ± 0.016 

0.18 % 

1.25 % 

-0.292  0.422 

-0.626 

0.629 

ST_PREF_left 

0.215 ± 0.009  0.216 ± 0.008  0.220 ± 0.015 

0.221 ± 0.016 

0.29 % 

0.41 % 

-0.323  0.422 

-0.391 

0.629 

ST_PREF_right 

0.214 ± 0.008  0.214 ± 0.008  0.217 ± 0.012 

0.218 ± 0.016 

0.03 % 

0.69 % 

0.136 

0.441 

-0.494 

0.629 

ST_PREM_left 

0.207 ± 0.009  0.208 ± 0.009  0.213 ± 0.017 

0.215 ± 0.015 

0.46 % 

1.27 % 

-0.554  0.422 

-0.813 

0.629 

ST_PREM_right 

0.216 ± 0.008  0.216 ± 0.009  0.218 ± 0.016 

0.221 ± 0.018 

0.26 % 

1.1 % 

-0.693  0.422 

-0.461 

0.629 

T_OCC_left 

0.220 ± 0.011  0.220 ± 0.011  0.221 ± 0.006 

0.218 ± 0.027 

-0.02 % 

-1.35 % 

0.224 

0.422 

T_OCC_right 

0.227 ± 0.011  0.227 ± 0.011  0.226 ± 0.009 

0.224 ± 0.019 

-0.04 % 

-0.68 % 

0.572 

0.422 

0.491 

0.331 

0.629 

0.629 

T_PAR_left 

0.216 ± 0.010  0.217 ± 0.009  0.214 ± 0.016 

0.217 ± 0.017 

0.36 % 

1.36 % 

-0.332  0.422 

-0.693 

0.629 

T_PAR_right 

0.228 ± 0.011  0.228 ± 0.010  0.226 ± 0.015 

0.228 ± 0.016 

0.08 % 

1.05 % 

0.194 

0.425 

-0.496 

0.629 

T_POSTC_left 

0.227 ± 0.010  0.228 ± 0.012  0.226 ± 0.021 

0.233 ± 0.019 

0.32 % 

2.76 % 

-0.358  0.422 

-1.043 

0.629 

T_POSTC_right 

0.239 ± 0.013  0.238 ± 0.015  0.235 ± 0.025 

0.238 ± 0.024 

-0.28 % 

1.38 % 

0.425 

0.422 

-0.442 

0.629 

T_PREC_left 

0.225 ± 0.009  0.226 ± 0.012  0.231 ± 0.019 

0.235 ± 0.017 

0.56 % 

1.79 % 

-1.048  0.422 

-0.945 

0.629 

T_PREC_right 

0.230 ± 0.010  0.231 ± 0.011  0.231 ± 0.018 

0.234 ± 0.016 

0.45 % 

1.47 % 

-0.801  0.422 

-0.725 

0.629 

T_PREF_left 

0.215 ± 0.009  0.216 ± 0.010  0.220 ± 0.016 

0.223 ± 0.017 

0.49 % 

1.05 % 

-0.751  0.422 

-0.693 

0.629 

T_PREF_right 

0.209 ± 0.008  0.210 ± 0.009  0.212 ± 0.013 

0.214 ± 0.015 

0.39 % 

1.22 % 

-0.892  0.422 

-0.905 

0.629 

T_PREM_left 

0.212 ± 0.009  0.214 ± 0.011  0.219 ± 0.017 

0.224 ± 0.016 

0.66 % 

2.24 % 

-0.837  0.422 

-1.182 

0.629 

T_PREM_right 

0.217 ± 0.010  0.219 ± 0.011  0.220 ± 0.020 

0.224 ± 0.019 

0.72 % 

1.63 % 

-1.279  0.422 

-0.771 

0.629 

UF_left 

0.221 ± 0.012  0.219 ± 0.014  0.224 ± 0.014 

0.220 ± 0.026 

-1.12 % 

-1.77 % 

1.708 

0.422 

0.497 

0.629 

0.215 ± 0.010  0.214 ± 0.015  0.218 ± 0.011 

UF_right 
Tables A.19- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

0.216 ± 0.028 

-0.82 % 

-0.92 % 

0.422 

1.328 

0.252 

0.629 

 298 

 
 
 
 
 
 
 
 
 
 
 
Table A.20 Results of Post Hoc Tract Specific Comparison of Diffusion Kurtosis Imaging Mean 
Kurtosis 

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

mTBI 

Controls 

P1 

P2 

P1 

P2 

9.405e-04 ± 
2.940e-05 

9.296e-04 ± 
2.842e-05 

9.992e-04 ± 
4.547e-05 

1.011e-03 ± 
5.214e-05 

1.051e-03 ± 
3.639e-05 

1.027e-03 ± 
5.777e-05 

9.983e-04 ± 
3.397e-05 

1.023e-03 ± 
4.521e-05 

1.015e-03 ± 
3.549e-05 

1.004e-03 ± 
3.478e-05 

9.667e-04 ± 
4.284e-05 

9.930e-04 ± 
7.692e-05 

9.882e-04 ± 
3.531e-05 

9.343e-04 ± 
2.881e-05 

9.228e-04 ± 
3.196e-05 

1.005e-03 ± 
3.049e-05 

9.612e-04 ± 
3.090e-05 

1.005e-03 ± 
2.968e-05 

9.983e-04 ± 
3.213e-05 

2.084e-03 ± 
3.622e-04 

2.055e-03 ± 
3.465e-04 

8.986e-04 ± 
8.235e-05 

8.806e-04 ± 
6.453e-05 

9.595e-04 ± 
3.503e-05 

9.506e-04 ± 
4.005e-05 

9.724e-04 ± 
5.191e-05 

9.614e-04 ± 
5.392e-05 

9.323e-04 ± 
3.288e-05 

9.242e-04 ± 
2.501e-05 

9.917e-04 ± 
5.552e-05 

9.990e-04 ± 
5.000e-05 

1.045e-03 ± 
5.522e-05 

1.019e-03 ± 
6.614e-05 

9.973e-04 ± 
4.112e-05 

1.026e-03 ± 
6.617e-05 

1.021e-03 ± 
4.994e-05 

1.015e-03 ± 
5.405e-05 

9.703e-04 ± 
3.623e-05 

9.924e-04 ± 
5.292e-05 

9.895e-04 ± 
3.479e-05 

9.320e-04 ± 
2.868e-05 

9.186e-04 ± 
2.781e-05 

1.002e-03 ± 
4.362e-05 

9.573e-04 ± 
3.144e-05 

1.003e-03 ± 
4.230e-05 

9.962e-04 ± 
3.487e-05 

2.053e-03 ± 
3.602e-04 

2.035e-03 ± 
3.660e-04 

8.888e-04 ± 
5.192e-05 

8.727e-04 ± 
4.092e-05 

9.568e-04 ± 
3.508e-05 

9.457e-04 ± 
3.215e-05 

9.619e-04 ± 
4.314e-05 

9.529e-04 ± 
4.118e-05 

9.049e-04 ± 
3.749e-05 

9.134e-04 ± 
3.055e-05 

9.651e-04 ± 
5.828e-05 

9.922e-04 ± 
7.752e-05 

9.882e-04 ± 
4.902e-05 

1.018e-03 ± 
9.343e-05 

9.747e-04 ± 
4.772e-05 

9.953e-04 ± 
6.128e-05 

9.975e-04 ± 
5.900e-05 

9.915e-04 ± 
6.630e-05 

9.569e-04 ± 
3.907e-05 

9.851e-04 ± 
4.702e-05 

9.744e-04 ± 
4.187e-05 

9.146e-04 ± 
5.173e-05 

9.050e-04 ± 
4.795e-05 

9.677e-04 ± 
5.499e-05 

9.326e-04 ± 
4.490e-05 

9.738e-04 ± 
7.679e-05 

9.747e-04 ± 
7.129e-05 

2.071e-03 ± 
3.822e-04 

2.050e-03 ± 
3.706e-04 

8.881e-04 ± 
6.794e-05 

8.802e-04 ± 
7.328e-05 

9.338e-04 ± 
2.693e-05 

9.348e-04 ± 
3.207e-05 

9.423e-04 ± 
3.590e-05 

9.451e-04 ± 
4.685e-05 

8.943e-04 ± 
4.150e-05 

9.025e-04 ± 
3.455e-05 

9.416e-04 ± 
6.102e-05 

9.710e-04 ± 
7.387e-05 

9.673e-04 ± 
5.410e-05 

9.792e-04 ± 
6.515e-05 

9.563e-04 ± 
5.529e-05 

9.825e-04 ± 
8.887e-05 

1.003e-03 ± 
1.205e-04 

9.959e-04 ± 
1.314e-04 

9.626e-04 ± 
7.973e-05 

9.946e-04 ± 
8.659e-05 

9.700e-04 ± 
7.486e-05 

9.169e-04 ± 
9.080e-05 

8.982e-04 ± 
7.526e-05 

9.620e-04 ± 
9.214e-05 

9.427e-04 ± 
9.243e-05 

9.531e-04 ± 
8.568e-05 

9.528e-04 ± 
7.834e-05 

2.052e-03 ± 
4.225e-04 

1.971e-03 ± 
4.431e-04 

8.751e-04 ± 
8.569e-05 

8.455e-04 ± 
9.221e-05 

9.277e-04 ± 
4.550e-05 

9.272e-04 ± 
3.808e-05 

9.315e-04 ± 
5.478e-05 

9.341e-04 ± 
6.133e-05 

 299 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.87 % 

-1.18 % 

2.102 

0.124 

1.141 

0.741 

-0.58 % 

-1.19 % 

1.999 

0.124 

1.390 

0.732 

-0.75 % 

-2.43 % 

1.195 

0.149 

1.755 

0.732 

-1.2 % 

-2.13 % 

2.129 

0.124 

1.524 

0.732 

-0.58 % 

-2.11 % 

0.671 

0.206 

1.173 

0.741 

-0.81 % 

-3.77 % 

1.041 

0.172 

2.976 

0.319 

-0.1 % 

-1.89 % 

0.067 

0.235 

1.296 

0.732 

0.35 % 

-1.29 % 

-0.499 

0.206 

0.617 

0.872 

0.61 % 

0.6 % 

-0.999 

0.172 

-0.075 

0.889 

1.13 % 

0.44 % 

-1.769 

0.145 

-0.066 

0.889 

0.37 % 

0.59 % 

-1.251 

0.149 

-0.190 

0.885 

-0.06 % 

0.96 % 

-0.122 

0.232 

-0.215 

0.884 

0.13 % 

-0.45 % 

-0.653 

0.206 

0.391 

0.872 

-0.25 % 

0.24 % 

0.386 

0.206 

-0.062 

0.889 

-0.46 % 

-0.75 % 

0.684 

0.206 

0.357 

0.872 

-0.34 % 

-0.59 % 

0.606 

0.206 

0.339 

0.872 

-0.4 % 

1.08 % 

0.512 

0.206 

-0.431 

0.872 

-0.18 % 

-2.13 % 

0.133 

0.232 

0.917 

0.807 

-0.21 % 

-2.25 % 

0.170 

0.232 

1.219 

0.741 

-1.49 % 

-0.92 % 

1.356 

0.149 

0.449 

0.872 

-1.01 % 

-3.89 % 

0.563 

0.206 

1.460 

0.732 

-1.08 % 

-1.47 % 

0.836 

0.194 

0.537 

0.872 

-0.9 % 

-3.94 % 

0.802 

0.198 

1.528 

0.732 

-0.28 % 

-0.66 % 

0.598 

0.206 

0.902 

0.807 

-0.51 % 

-0.82 % 

1.023 

0.172 

2.739 

0.319 

-1.07 % 

-1.15 % 

2.123 

0.124 

1.104 

0.741 

-0.88 % 

-1.16 % 

1.570 

0.149 

2.178 

0.512 

 
 
 
 
 
 
 
 
Table A.20 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

8.708e-04 ± 
6.393e-05 

9.510e-04 ± 
3.528e-05 

9.415e-04 ± 
3.877e-05 

9.945e-04 ± 
5.449e-05 

9.831e-04 ± 
6.108e-05 

9.991e-04 ± 
3.297e-05 

9.794e-04 ± 
3.744e-05 

9.493e-04 ± 
3.634e-05 

9.361e-04 ± 
2.950e-05 

9.434e-04 ± 
3.034e-05 

9.267e-04 ± 
2.895e-05 

9.559e-04 ± 
3.748e-05 

9.449e-04 ± 
3.611e-05 

9.992e-04 ± 
5.026e-05 

9.798e-04 ± 
4.602e-05 

9.481e-04 ± 
2.894e-05 

9.235e-04 ± 
2.917e-05 

9.471e-04 ± 
4.708e-05 

9.540e-04 ± 
4.446e-05 

9.709e-04 ± 
4.304e-05 

9.519e-04 ± 
4.755e-05 

9.781e-04 ± 
3.716e-05 

9.555e-04 ± 
3.860e-05 

9.937e-04 ± 
3.872e-05 

9.563e-04 ± 
3.757e-05 

9.860e-04 ± 
3.607e-05 

9.556e-04 ± 
3.735e-05 

8.685e-04 ± 
4.820e-05 

9.494e-04 ± 
3.444e-05 

9.415e-04 ± 
3.484e-05 

9.898e-04 ± 
5.405e-05 

9.795e-04 ± 
5.414e-05 

1.003e-03 ± 
3.879e-05 

9.816e-04 ± 
3.672e-05 

9.421e-04 ± 
4.372e-05 

9.290e-04 ± 
3.024e-05 

9.396e-04 ± 
3.632e-05 

9.254e-04 ± 
2.677e-05 

9.506e-04 ± 
3.974e-05 

9.399e-04 ± 
3.350e-05 

1.002e-03 ± 
5.790e-05 

9.771e-04 ± 
5.018e-05 

9.418e-04 ± 
4.814e-05 

9.162e-04 ± 
3.585e-05 

9.451e-04 ± 
5.533e-05 

9.478e-04 ± 
4.387e-05 

9.666e-04 ± 
4.266e-05 

9.476e-04 ± 
4.183e-05 

9.795e-04 ± 
3.927e-05 

9.552e-04 ± 
3.533e-05 

9.907e-04 ± 
5.073e-05 

9.534e-04 ± 
3.951e-05 

9.786e-04 ± 
5.040e-05 

9.469e-04 ± 
3.475e-05 

8.572e-04 ± 
6.320e-05 

9.384e-04 ± 
3.773e-05 

9.359e-04 ± 
4.410e-05 

9.744e-04 ± 
4.405e-05 

9.841e-04 ± 
5.234e-05 

9.932e-04 ± 
5.520e-05 

9.688e-04 ± 
5.193e-05 

9.138e-04 ± 
5.534e-05 

9.207e-04 ± 
7.218e-05 

9.186e-04 ± 
6.050e-05 

9.196e-04 ± 
3.960e-05 

9.367e-04 ± 
5.163e-05 

9.310e-04 ± 
4.893e-05 

1.011e-03 ± 
1.104e-04 

9.788e-04 ± 
8.196e-05 

8.943e-04 ± 
5.409e-05 

8.993e-04 ± 
6.302e-05 

9.020e-04 ± 
4.637e-05 

9.270e-04 ± 
5.471e-05 

9.416e-04 ± 
2.495e-05 

9.384e-04 ± 
3.310e-05 

9.692e-04 ± 
4.808e-05 

9.441e-04 ± 
4.045e-05 

9.698e-04 ± 
6.816e-05 

9.484e-04 ± 
6.615e-05 

9.413e-04 ± 
5.738e-05 

9.327e-04 ± 
5.116e-05 

8.784e-04 ± 
8.677e-05 

9.319e-04 ± 
5.611e-05 

9.289e-04 ± 
5.616e-05 

9.733e-04 ± 
8.667e-05 

9.772e-04 ± 
7.327e-05 

9.883e-04 ± 
9.957e-05 

9.708e-04 ± 
9.200e-05 

9.047e-04 ± 
7.949e-05 

9.000e-04 ± 
5.096e-05 

9.067e-04 ± 
6.975e-05 

9.110e-04 ± 
5.053e-05 

9.242e-04 ± 
6.602e-05 

9.181e-04 ± 
5.575e-05 

9.988e-04 ± 
1.340e-04 

9.684e-04 ± 
1.147e-04 

9.001e-04 ± 
1.305e-04 

8.863e-04 ± 
8.140e-05 

8.959e-04 ± 
6.989e-05 

9.033e-04 ± 
3.635e-05 

9.403e-04 ± 
5.582e-05 

9.355e-04 ± 
4.555e-05 

9.649e-04 ± 
8.628e-05 

9.445e-04 ± 
7.279e-05 

9.574e-04 ± 
1.078e-04 

9.513e-04 ± 
1.081e-04 

9.381e-04 ± 
9.289e-05 

9.306e-04 ± 
7.729e-05 

 300 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-
score 
mTBI 

q-
value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.27 % 

2.47 % 

0.194 

0.232 

-0.701 

0.872 

-0.18 % 

-0.69 % 

0.090 

0.234 

0.585 

0.872 

0 % 

-0.75 % 

-0.557  0.206 

0.624 

0.872 

-0.47 % 

-0.12 % 

0.643 

0.206 

0.113 

0.889 

-0.37 % 

-0.7 % 

0.116 

0.232 

1.280 

0.732 

0.39 % 

-0.49 % 

-1.218  0.149 

0.302 

0.872 

0.22 % 

0.2 % 

-0.967  0.173 

-0.104 

0.889 

-0.75 % 

-0.99 % 

1.251 

0.149 

0.379 

0.872 

-0.75 % 

-2.25 % 

1.880 

0.128 

1.362 

0.732 

-0.4 % 

-1.29 % 

1.003 

0.172 

0.644 

0.872 

-0.15 % 

-0.94 % 

0.387 

0.206 

0.659 

0.872 

-0.56 % 

-1.33 % 

1.444 

0.149 

0.970 

0.807 

-0.54 % 

-1.39 % 

1.310 

0.149 

1.189 

0.741 

0.29 % 

-1.24 % 

-0.553  0.206 

0.471 

0.872 

-0.27 % 

-1.06 % 

0.427 

0.206 

0.523 

0.872 

-0.66 % 

0.65 % 

1.102 

0.167 

-0.242 

0.884 

-0.8 % 

-1.45 % 

1.433 

0.149 

0.638 

0.872 

-0.21 % 

-0.67 % 

0.172 

0.232 

0.444 

0.872 

-0.65 % 

-2.56 % 

1.302 

0.149 

2.638 

0.319 

-0.44 % 

-0.14 % 

0.878 

0.188 

0.291 

0.872 

-0.45 % 

-0.3 % 

0.391 

0.206 

2.671 

0.319 

0.14 % 

-0.45 % 

-0.698  0.206 

0.371 

0.872 

-0.03 % 

0.04 % 

-0.463  0.206 

0.053 

0.889 

-0.3 % 

-1.27 % 

0.514 

0.206 

0.567 

0.872 

-0.31 % 

0.31 % 

0.429 

0.206 

0.015 

0.898 

-0.75 % 

-0.34 % 

1.449 

0.149 

0.276 

0.872 

-0.91 % 

-0.22 % 

1.549 

0.149 

0.325 

0.872 

 
 
 
 
 
 
 
Table A.20 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

9.640e-04 ± 
3.279e-05 

9.622e-04 ± 
3.418e-05 

9.727e-04 ± 
3.862e-05 

9.570e-04 ± 
3.336e-05 

1.001e-03 ± 
5.633e-05 

9.842e-04 ± 
6.157e-05 

9.924e-04 ± 
4.415e-05 

9.798e-04 ± 
4.823e-05 

9.904e-04 ± 
3.971e-05 

9.554e-04 ± 
4.229e-05 

9.895e-04 ± 
3.827e-05 

9.530e-04 ± 
4.062e-05 

9.875e-04 ± 
3.675e-05 

9.944e-04 ± 
4.383e-05 

9.990e-04 ± 
4.493e-05 

9.810e-04 ± 
4.743e-05 

9.459e-04 ± 
3.425e-05 

9.729e-04 ± 
3.170e-05 

9.582e-04 ± 
4.179e-05 

9.550e-04 ± 
3.441e-05 

9.648e-04 ± 
5.527e-05 

9.475e-04 ± 
3.796e-05 

9.958e-04 ± 
5.548e-05 

9.801e-04 ± 
5.434e-05 

9.945e-04 ± 
4.689e-05 

9.789e-04 ± 
4.424e-05 

9.908e-04 ± 
5.222e-05 

9.565e-04 ± 
4.577e-05 

9.824e-04 ± 
5.481e-05 

9.440e-04 ± 
3.812e-05 

9.809e-04 ± 
4.894e-05 

9.846e-04 ± 
4.471e-05 

9.901e-04 ± 
6.565e-05 

9.687e-04 ± 
5.084e-05 

9.485e-04 ± 
3.537e-05 

9.701e-04 ± 
3.521e-05 

9.262e-04 ± 
5.308e-05 

9.342e-04 ± 
5.387e-05 

9.254e-04 ± 
5.624e-05 

9.252e-04 ± 
6.306e-05 

9.813e-04 ± 
4.437e-05 

9.846e-04 ± 
5.332e-05 

9.928e-04 ± 
6.218e-05 

9.773e-04 ± 
6.115e-05 

9.876e-04 ± 
8.681e-05 

9.661e-04 ± 
9.372e-05 

9.537e-04 ± 
6.981e-05 

9.369e-04 ± 
6.181e-05 

9.550e-04 ± 
6.698e-05 

9.749e-04 ± 
7.562e-05 

9.630e-04 ± 
7.541e-05 

9.640e-04 ± 
1.017e-04 

9.204e-04 ± 
4.060e-05 

9.535e-04 ± 
5.527e-05 

9.136e-04 ± 
6.012e-05 

9.133e-04 ± 
4.582e-05 

8.975e-04 ± 
4.967e-05 

9.101e-04 ± 
5.958e-05 

9.800e-04 ± 
8.538e-05 

9.783e-04 ± 
7.429e-05 

9.876e-04 ± 
1.092e-04 

9.748e-04 ± 
9.990e-05 

9.722e-04 ± 
1.256e-04 

9.680e-04 ± 
1.417e-04 

9.490e-04 ± 
1.109e-04 

9.331e-04 ± 
9.125e-05 

9.383e-04 ± 
7.597e-05 

9.510e-04 ± 
7.470e-05 

9.316e-04 ± 
8.453e-05 

9.429e-04 ± 
9.568e-05 

9.080e-04 ± 
4.121e-05 

9.301e-04 ± 
4.556e-05 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-
score 
mTBI 

q-
value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.6 % 

-1.36 % 

1.416 

0.149 

0.864 

0.823 

-0.75 % 

-2.24 % 

2.212 

0.124 

1.962 

0.654 

-0.81 % 

-3.02 % 

1.538 

0.149 

2.311 

0.479 

-0.99 % 

-1.64 % 

2.792 

0.124 

1.341 

0.732 

-0.53 % 

-0.14 % 

0.757 

0.205 

0.142 

0.889 

-0.41 % 

-0.63 % 

0.210 

0.232 

1.273 

0.732 

0.21 % 

-0.53 % 

-0.635  0.206 

0.323 

0.872 

-0.1 % 

-0.26 % 

-0.153  0.232 

0.210 

0.884 

0.04 % 

-1.56 % 

-0.034  0.238 

0.571 

0.872 

0.12 % 

0.19 % 

-0.225  0.232 

0.045 

0.889 

-0.72 % 

-0.49 % 

1.299 

0.149 

0.276 

0.872 

-0.94 % 

-0.41 % 

1.354 

0.149 

0.330 

0.872 

-0.67 % 

-1.76 % 

1.239 

0.149 

0.904 

0.807 

-0.98 % 

-2.44 % 

1.951 

0.124 

1.522 

0.732 

-0.9 % 

-3.26 % 

1.242 

0.149 

1.582 

0.732 

-1.26 % 

-2.19 % 

1.956 

0.124 

1.089 

0.741 

0.28 % 

-1.35 % 

-0.957  0.173 

1.023 

0.785 

-0.29 % 

-2.46 % 

0.485 

0.206 

1.455 

0.732 

Tables A.20- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 301 

 
 
 
 
 
 
 
 
 
 
 
Table A.21 Results of Post Hoc Tract Specific Comparison of Diffusion Kurtosis Imaging Axial 
Kurtosis 

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

mTBI 

Controls 

P1 

P2 

P1 

P2 

1.137e-03 ± 
2.870e-05 

1.126e-03 ± 
2.878e-05 

1.207e-03 ± 
4.550e-05 

1.216e-03 ± 
5.206e-05 

1.285e-03 ± 
3.787e-05 

1.318e-03 ± 
6.172e-05 

1.238e-03 ± 
3.436e-05 

1.267e-03 ± 
4.850e-05 

1.264e-03 ± 
3.244e-05 

1.260e-03 ± 
3.440e-05 

1.204e-03 ± 
3.913e-05 

1.240e-03 ± 
7.382e-05 

1.225e-03 ± 
3.237e-05 

1.148e-03 ± 
3.020e-05 

1.136e-03 ± 
3.228e-05 

1.278e-03 ± 
3.066e-05 

1.232e-03 ± 
3.136e-05 

1.265e-03 ± 
3.123e-05 

1.260e-03 ± 
3.421e-05 

2.431e-03 ± 
3.183e-04 

2.408e-03 ± 
3.038e-04 

1.081e-03 ± 
8.088e-05 

1.051e-03 ± 
6.341e-05 

1.181e-03 ± 
3.524e-05 

1.168e-03 ± 
4.034e-05 

1.195e-03 ± 
5.468e-05 

1.179e-03 ± 
5.903e-05 

1.127e-03 ± 
3.362e-05 

1.119e-03 ± 
2.558e-05 

1.199e-03 ± 
5.503e-05 

1.203e-03 ± 
4.988e-05 

1.276e-03 ± 
5.025e-05 

1.306e-03 ± 
7.864e-05 

1.235e-03 ± 
4.059e-05 

1.267e-03 ± 
6.164e-05 

1.267e-03 ± 
4.850e-05 

1.268e-03 ± 
5.256e-05 

1.207e-03 ± 
3.687e-05 

1.239e-03 ± 
5.439e-05 

1.225e-03 ± 
3.325e-05 

1.147e-03 ± 
2.997e-05 

1.132e-03 ± 
2.951e-05 

1.272e-03 ± 
4.403e-05 

1.226e-03 ± 
3.289e-05 

1.263e-03 ± 
4.285e-05 

1.256e-03 ± 
3.649e-05 

2.393e-03 ± 
3.217e-04 

2.381e-03 ± 
3.238e-04 

1.069e-03 ± 
5.278e-05 

1.040e-03 ± 
4.193e-05 

1.177e-03 ± 
3.770e-05 

1.162e-03 ± 
3.385e-05 

1.181e-03 ± 
5.091e-05 

1.169e-03 ± 
4.907e-05 

1.095e-03 ± 
3.274e-05 

1.105e-03 ± 
3.145e-05 

1.169e-03 ± 
5.415e-05 

1.193e-03 ± 
7.617e-05 

1.223e-03 ± 
4.878e-05 

1.310e-03 ± 
9.903e-05 

1.207e-03 ± 
4.365e-05 

1.232e-03 ± 
5.686e-05 

1.234e-03 ± 
5.206e-05 

1.235e-03 ± 
5.934e-05 

1.185e-03 ± 
3.034e-05 

1.227e-03 ± 
5.163e-05 

1.204e-03 ± 
3.478e-05 

1.123e-03 ± 
4.682e-05 

1.112e-03 ± 
3.971e-05 

1.226e-03 ± 
4.775e-05 

1.192e-03 ± 
4.279e-05 

1.221e-03 ± 
6.699e-05 

1.224e-03 ± 
6.438e-05 

2.407e-03 ± 
3.339e-04 

2.386e-03 ± 
3.260e-04 

1.059e-03 ± 
5.528e-05 

1.039e-03 ± 
6.510e-05 

1.152e-03 ± 
2.899e-05 

1.150e-03 ± 
3.763e-05 

1.163e-03 ± 
4.061e-05 

1.163e-03 ± 
5.609e-05 

1.082e-03 ± 
3.656e-05 

1.089e-03 ± 
2.307e-05 

1.143e-03 ± 
5.492e-05 

1.168e-03 ± 
6.844e-05 

1.194e-03 ± 
3.581e-05 

1.255e-03 ± 
8.869e-05 

1.184e-03 ± 
4.207e-05 

1.214e-03 ± 
7.173e-05 

1.237e-03 ± 
1.013e-04 

1.236e-03 ± 
1.116e-04 

1.188e-03 ± 
6.570e-05 

1.229e-03 ± 
7.370e-05 

1.195e-03 ± 
6.046e-05 

1.121e-03 ± 
8.185e-05 

1.100e-03 ± 
6.918e-05 

1.218e-03 ± 
8.127e-05 

1.197e-03 ± 
7.733e-05 

1.197e-03 ± 
7.479e-05 

1.199e-03 ± 
6.964e-05 

2.369e-03 ± 
3.880e-04 

2.295e-03 ± 
4.230e-04 

1.043e-03 ± 
6.917e-05 

1.004e-03 ± 
9.658e-05 

1.139e-03 ± 
3.157e-05 

1.137e-03 ± 
3.810e-05 

1.142e-03 ± 
4.550e-05 

1.143e-03 ± 
7.127e-05 

 302 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.86 % 

-1.15 % 

2.577 

0.116 

1.638 

0.173 

-0.68 % 

-1.42 % 

2.893 

0.085 

2.740 

0.096 

-0.7 % 

-2.24 % 

1.423 

0.221 

2.145 

0.151 

-1.1 % 

-2.09 % 

2.379 

0.132 

1.863 

0.16 

-0.69 % 

-2.44 % 

1.168 

0.283 

2.154 

0.151 

-0.85 % 

-4.21 % 

1.293 

0.244 

2.387 

0.142 

-0.25 % 

-1.97 % 

0.621 

0.398 

2.171 

0.151 

0.02 % 

-1.48 % 

-0.062 

0.556 

0.967 

0.353 

0.23 % 

0.19 % 

-0.529 

0.421 

0.014 

0.63 % 

0.08 % 

-1.372 

0.234 

-0.013 

0.22 % 

0.21 % 

-1.186 

0.283 

-0.084 

0.5 

0.5 

0.5 

-0.13 % 

0.18 % 

0.049 

0.556 

0.462 

0.412 

-0.02 % 

-0.75 % 

-0.244 

0.512 

0.865 

0.379 

-0.06 % 

-0.15 % 

-0.017 

0.556 

0.065 

0.5 

-0.39 % 

-1.08 % 

0.668 

0.393 

0.633 

0.398 

-0.4 % 

-0.69 % 

0.982 

0.319 

0.566 

0.4 

-0.48 % 

0.46 % 

0.973 

0.319 

-0.285 

0.449 

-0.23 % 

-1.98 % 

0.459 

0.44 

1.311 

0.239 

-0.29 % 

-2.09 % 

0.569 

0.416 

1.749 

0.169 

-1.57 % 

-1.56 % 

1.864 

0.198 

0.644 

0.398 

-1.11 % 

-3.82 % 

1.035 

0.317 

1.411 

0.218 

-1.1 % 

-1.47 % 

1.064 

0.316 

0.831 

0.379 

-1.08 % 

-3.33 % 

1.324 

0.239 

1.524 

0.194 

-0.31 % 

-1.11 % 

0.932 

0.331 

3.277 

0.096 

-0.54 % 

-1.11 % 

1.528 

0.198 

2.962 

0.096 

-1.1 % 

-1.82 % 

2.338 

0.132 

2.029 

0.155 

-0.88 % 

-1.66 % 

1.709 

0.198 

2.009 

0.155 

 
 
 
 
 
 
 
 
Table A.21 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

1.071e-03 ± 
6.466e-05 

1.066e-03 ± 
5.086e-05 

1.047e-03 ± 
5.328e-05 

1.066e-03 ± 
6.899e-05 

1.155e-03 ± 
3.379e-05 

1.153e-03 ± 
3.526e-05 

1.136e-03 ± 
3.268e-05 

1.128e-03 ± 
5.063e-05 

1.148e-03 ± 
3.761e-05 

1.146e-03 ± 
3.513e-05 

1.136e-03 ± 
3.777e-05 

1.125e-03 ± 
4.764e-05 

1.229e-03 ± 
5.251e-05 

1.223e-03 ± 
5.395e-05 

1.205e-03 ± 
4.734e-05 

1.196e-03 ± 
6.880e-05 

1.221e-03 ± 
5.889e-05 

1.216e-03 ± 
5.357e-05 

1.219e-03 ± 
5.695e-05 

1.206e-03 ± 
7.741e-05 

1.244e-03 ± 
3.416e-05 

1.247e-03 ± 
3.929e-05 

1.227e-03 ± 
4.715e-05 

1.220e-03 ± 
9.075e-05 

1.231e-03 ± 
3.839e-05 

1.231e-03 ± 
3.888e-05 

1.212e-03 ± 
4.598e-05 

1.212e-03 ± 
8.075e-05 

1.179e-03 ± 
3.726e-05 

1.170e-03 ± 
4.364e-05 

1.130e-03 ± 
4.542e-05 

1.120e-03 ± 
6.357e-05 

1.164e-03 ± 
3.064e-05 

1.154e-03 ± 
3.182e-05 

1.134e-03 ± 
6.193e-05 

1.115e-03 ± 
4.157e-05 

1.143e-03 ± 
3.106e-05 

1.136e-03 ± 
3.644e-05 

1.110e-03 ± 
4.811e-05 

1.097e-03 ± 
6.196e-05 

1.128e-03 ± 
2.968e-05 

1.123e-03 ± 
2.680e-05 

1.111e-03 ± 
3.722e-05 

1.096e-03 ± 
3.389e-05 

1.143e-03 ± 
3.460e-05 

1.138e-03 ± 
3.768e-05 

1.118e-03 ± 
4.439e-05 

1.107e-03 ± 
6.121e-05 

1.131e-03 ± 
3.491e-05 

1.126e-03 ± 
3.241e-05 

1.111e-03 ± 
4.475e-05 

1.098e-03 ± 
4.853e-05 

1.190e-03 ± 
4.919e-05 

1.192e-03 ± 
5.473e-05 

1.194e-03 ± 
9.878e-05 

1.180e-03 ± 
1.248e-04 

1.176e-03 ± 
4.418e-05 

1.173e-03 ± 
4.698e-05 

1.168e-03 ± 
7.384e-05 

1.158e-03 ± 
1.053e-04 

1.201e-03 ± 
3.187e-05 

1.194e-03 ± 
4.787e-05 

1.135e-03 ± 
5.223e-05 

1.137e-03 ± 
1.172e-04 

1.177e-03 ± 
3.361e-05 

1.168e-03 ± 
3.931e-05 

1.142e-03 ± 
5.843e-05 

1.127e-03 ± 
7.560e-05 

1.179e-03 ± 
5.018e-05 

1.175e-03 ± 
6.145e-05 

1.132e-03 ± 
5.240e-05 

1.118e-03 ± 
5.352e-05 

1.192e-03 ± 
4.641e-05 

1.184e-03 ± 
5.045e-05 

1.162e-03 ± 
5.716e-05 

1.130e-03 ± 
3.634e-05 

1.206e-03 ± 
4.186e-05 

1.201e-03 ± 
4.476e-05 

1.171e-03 ± 
2.893e-05 

1.163e-03 ± 
3.766e-05 

ST_OCC_right 

1.184e-03 ± 
4.732e-05 

1.179e-03 ± 
4.336e-05 

1.167e-03 ± 
4.299e-05 

1.160e-03 ± 
4.903e-05 

ST_PAR_left 

ST_PAR_right 

1.196e-03 ± 
3.686e-05 

1.197e-03 ± 
3.992e-05 

1.179e-03 ± 
3.994e-05 

1.173e-03 ± 
7.820e-05 

1.180e-03 ± 
3.836e-05 

1.178e-03 ± 
3.605e-05 

1.161e-03 ± 
3.700e-05 

1.160e-03 ± 
6.385e-05 

ST_POSTC_left 

1.221e-03 ± 
4.183e-05 

1.217e-03 ± 
5.251e-05 

1.186e-03 ± 
6.182e-05 

1.172e-03 ± 
1.026e-04 

ST_POSTC_right 

1.188e-03 ± 
3.885e-05 

1.183e-03 ± 
3.930e-05 

1.172e-03 ± 
6.255e-05 

1.173e-03 ± 
9.963e-05 

ST_PREC_left 

1.214e-03 ± 
3.774e-05 

1.206e-03 ± 
5.141e-05 

1.159e-03 ± 
5.239e-05 

1.153e-03 ± 
8.677e-05 

ST_PREC_right 

1.185e-03 ± 
3.839e-05 

1.175e-03 ± 
3.579e-05 

1.155e-03 ± 
5.168e-05 

1.150e-03 ± 
7.155e-05 

 303 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.44 % 

1.75 % 

0.466 

0.44 

-0.742 

0.379 

-0.15 % 

-0.69 % 

0.014 

0.556 

0.795 

0.379 

-0.14 % 

-0.91 % 

-0.118 

0.556 

1.115 

0.294 

-0.5 % 

-0.72 % 

0.912 

0.333 

0.778 

0.379 

-0.36 % 

-1.06 % 

0.296 

0.496 

1.690 

0.171 

0.23 % 

-0.55 % 

-1.135 

0.29 

0.443 

0.412 

0.07 % 

0 % 

-0.750 

0.373 

-0.011 

-0.75 % 

-0.85 % 

1.547 

0.198 

0.562 

0.5 

0.4 

-0.85 % 

-1.74 % 

2.834 

0.085 

1.869 

0.16 

-0.58 % 

-1.18 % 

1.967 

0.198 

0.819 

0.379 

-0.43 % 

-1.37 % 

1.547 

0.198 

1.842 

0.16 

-0.44 % 

-0.95 % 

1.522 

0.198 

0.897 

0.378 

-0.5 % 

-1.22 % 

1.701 

0.198 

1.492 

0.198 

0.23 % 

-1.18 % 

-0.646 

0.393 

0.560 

-0.23 % 

-0.89 % 

0.553 

0.416 

0.601 

0.4 

0.4 

-0.63 % 

0.16 % 

1.437 

0.221 

-0.142 

0.497 

-0.83 % 

-1.34 % 

1.899 

0.198 

0.753 

0.379 

-0.35 % 

-1.2 % 

0.652 

0.393 

1.591 

0.18 

-0.69 % 

-2.8 % 

1.721 

0.198 

2.665 

0.096 

-0.41 % 

-0.71 % 

1.026 

0.317 

1.703 

0.171 

-0.48 % 

-0.62 % 

0.759 

0.373 

2.064 

0.155 

0.12 % 

-0.51 % 

-0.767 

0.373 

0.488 

0.412 

-0.15 % 

-0.1 % 

-0.057 

0.556 

0.181 

0.487 

-0.33 % 

-1.19 % 

0.750 

0.373 

0.644 

0.398 

-0.43 % 

0.1 % 

0.817 

0.371 

0.082 

0.5 

-0.72 % 

-0.55 % 

1.703 

0.198 

0.408 

0.412 

-0.87 % 

-0.37 % 

1.877 

0.198 

0.519 

0.411 

 
 
 
 
 
 
 
Table A.21 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_PREF_left 

1.183e-03 ± 
3.438e-05 

1.176e-03 ± 
4.390e-05 

1.138e-03 ± 
4.938e-05 

1.122e-03 ± 
5.255e-05 

ST_PREF_right 

1.180e-03 ± 
3.657e-05 

1.172e-03 ± 
3.706e-05 

1.146e-03 ± 
5.420e-05 

1.121e-03 ± 
4.259e-05 

ST_PREM_left 

1.182e-03 ± 
4.221e-05 

1.173e-03 ± 
5.867e-05 

1.126e-03 ± 
5.127e-05 

1.094e-03 ± 
4.357e-05 

ST_PREM_right 

1.175e-03 ± 
3.808e-05 

1.164e-03 ± 
4.155e-05 

1.135e-03 ± 
6.074e-05 

1.117e-03 ± 
5.415e-05 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

1.233e-03 ± 
5.412e-05 

1.226e-03 ± 
5.529e-05 

1.209e-03 ± 
4.722e-05 

1.200e-03 ± 
6.843e-05 

1.217e-03 ± 
5.928e-05 

1.212e-03 ± 
5.384e-05 

1.215e-03 ± 
5.855e-05 

1.203e-03 ± 
7.828e-05 

1.210e-03 ± 
4.214e-05 

1.211e-03 ± 
4.516e-05 

1.202e-03 ± 
5.434e-05 

1.195e-03 ± 
9.944e-05 

1.205e-03 ± 
4.604e-05 

1.203e-03 ± 
4.326e-05 

1.196e-03 ± 
5.554e-05 

1.192e-03 ± 
8.996e-05 

T_POSTC_left 

1.222e-03 ± 
4.028e-05 

1.221e-03 ± 
5.075e-05 

1.205e-03 ± 
7.790e-05 

1.189e-03 ± 
1.154e-04 

T_POSTC_right 

1.190e-03 ± 
4.058e-05 

1.189e-03 ± 
4.220e-05 

1.190e-03 ± 
8.110e-05 

1.190e-03 ± 
1.282e-04 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

1.221e-03 ± 
3.738e-05 

1.213e-03 ± 
5.310e-05 

1.175e-03 ± 
6.239e-05 

1.167e-03 ± 
1.014e-04 

1.182e-03 ± 
3.977e-05 

1.172e-03 ± 
3.713e-05 

1.157e-03 ± 
5.724e-05 

1.151e-03 ± 
8.315e-05 

1.208e-03 ± 
3.669e-05 

1.200e-03 ± 
4.826e-05 

1.168e-03 ± 
6.137e-05 

1.148e-03 ± 
6.800e-05 

1.209e-03 ± 
4.391e-05 

1.198e-03 ± 
4.404e-05 

1.183e-03 ± 
7.357e-05 

1.156e-03 ± 
6.869e-05 

1.217e-03 ± 
4.501e-05 

1.207e-03 ± 
6.484e-05 

1.173e-03 ± 
6.971e-05 

1.139e-03 ± 
7.681e-05 

T_PREM_right 

1.202e-03 ± 
4.657e-05 

1.189e-03 ± 
5.014e-05 

1.177e-03 ± 
9.451e-05 

1.153e-03 ± 
8.559e-05 

UF_left 

1.173e-03 ± 
3.496e-05 

1.173e-03 ± 
3.626e-05 

1.144e-03 ± 
4.304e-05 

1.123e-03 ± 
2.854e-05 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.56 % 

-1.46 % 

1.670 

0.198 

1.284 

0.242 

-0.73 % 

-2.19 % 

2.524 

0.116 

2.331 

0.142 

-0.75 % 

-2.84 % 

1.785 

0.198 

3.145 

0.096 

-0.9 % 

-1.59 % 

2.970 

0.085 

1.806 

0.16 

-0.54 % 

-0.73 % 

1.011 

0.317 

0.807 

0.379 

-0.39 % 

-0.99 % 

0.388 

0.465 

1.639 

0.173 

0.15 % 

-0.58 % 

-0.703 

0.388 

0.422 

0.412 

-0.16 % 

-0.33 % 

0.001 

0.556 

0.296 

0.449 

-0.08 % 

-1.33 % 

0.231 

0.512 

0.655 

0.398 

-0.07 % 

0.01 % 

0.032 

0.556 

0.082 

0.5 

-0.7 % 

-0.65 % 

1.628 

0.198 

0.405 

0.412 

-0.86 % 

-0.51 % 

1.627 

0.198 

0.480 

0.412 

-0.61 % 

-1.74 % 

1.524 

0.198 

1.193 

0.27 

-0.9 % 

-2.3 % 

2.231 

0.149 

1.815 

0.16 

-0.82 % 

-2.97 % 

1.533 

0.198 

1.945 

0.16 

-1.08 % 

-2.04 % 

2.181 

0.149 

1.333 

0.239 

0.02 % 

-1.9 % 

-0.376 

0.465 

2.735 

0.096 

UF_right 

1.196e-03 ± 
3.558e-05 

1.201e-03 ± 
3.376e-05 

1.147e-03 ± 
2.801e-05 
Tables A.21- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

1.178e-03 ± 
5.722e-05 

-2.66 % 

-0.47 % 

1.337 

0.239 

2.687 

0.096 

 304 

 
 
 
 
 
 
 
 
 
 
 
Table A.22 Results of Post Hoc Tract Specific Comparison of Diffusion Kurtosis Imaging Radial 
Kurtosis 

mTBI 

Controls 

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

P1 

P2 

P1 

P2 

8.422e-04 ± 
3.098e-05 

8.312e-04 ± 
2.900e-05 

8.952e-04 ± 
4.606e-05 

9.084e-04 ± 
5.261e-05 

9.338e-04 ± 
3.814e-05 

8.816e-04 ± 
5.782e-05 

8.787e-04 ± 
3.511e-05 

9.005e-04 ± 
4.544e-05 

8.905e-04 ± 
3.930e-05 

8.753e-04 ± 
3.782e-05 

8.480e-04 ± 
4.596e-05 

8.694e-04 ± 
7.903e-05 

8.698e-04 ± 
3.778e-05 

8.276e-04 ± 
3.046e-05 

8.160e-04 ± 
3.399e-05 

8.687e-04 ± 
3.196e-05 

8.258e-04 ± 
3.205e-05 

8.750e-04 ± 
3.036e-05 

8.673e-04 ± 
3.241e-05 

1.910e-03 ± 
3.849e-04 

1.879e-03 ± 
3.690e-04 

8.072e-04 ± 
8.363e-05 

7.954e-04 ± 
6.566e-05 

8.488e-04 ± 
3.565e-05 

8.417e-04 ± 
4.028e-05 

8.612e-04 ± 
5.113e-05 

8.526e-04 ± 
5.201e-05 

8.348e-04 ± 
3.313e-05 

8.269e-04 ± 
2.550e-05 

8.882e-04 ± 
5.615e-05 

8.970e-04 ± 
5.055e-05 

9.291e-04 ± 
5.964e-05 

8.747e-04 ± 
6.486e-05 

8.786e-04 ± 
4.244e-05 

9.058e-04 ± 
6.995e-05 

8.983e-04 ± 
5.244e-05 

8.883e-04 ± 
5.664e-05 

8.520e-04 ± 
3.726e-05 

8.693e-04 ± 
5.339e-05 

8.719e-04 ± 
3.652e-05 

8.245e-04 ± 
3.115e-05 

8.119e-04 ± 
2.911e-05 

8.661e-04 ± 
4.453e-05 

8.229e-04 ± 
3.218e-05 

8.736e-04 ± 
4.320e-05 

8.661e-04 ± 
3.546e-05 

1.883e-03 ± 
3.814e-04 

1.862e-03 ± 
3.889e-04 

7.985e-04 ± 
5.271e-05 

7.892e-04 ± 
4.163e-05 

8.466e-04 ± 
3.460e-05 

8.375e-04 ± 
3.205e-05 

8.522e-04 ± 
4.061e-05 

8.451e-04 ± 
3.895e-05 

8.099e-04 ± 
4.067e-05 

8.177e-04 ± 
3.143e-05 

8.629e-04 ± 
6.084e-05 

8.917e-04 ± 
7.865e-05 

8.705e-04 ± 
5.066e-05 

8.714e-04 ± 
9.258e-05 

8.584e-04 ± 
5.187e-05 

8.769e-04 ± 
6.721e-05 

8.790e-04 ± 
6.563e-05 

8.696e-04 ± 
7.291e-05 

8.428e-04 ± 
4.536e-05 

8.642e-04 ± 
4.553e-05 

8.597e-04 ± 
4.716e-05 

8.105e-04 ± 
5.637e-05 

8.015e-04 ± 
5.317e-05 

8.385e-04 ± 
6.011e-05 

8.031e-04 ± 
4.801e-05 

8.500e-04 ± 
8.280e-05 

8.499e-04 ± 
7.568e-05 

1.903e-03 ± 
4.074e-04 

1.882e-03 ± 
3.939e-04 

8.028e-04 ± 
7.506e-05 

8.009e-04 ± 
7.779e-05 

8.250e-04 ± 
2.747e-05 

8.273e-04 ± 
3.026e-05 

8.321e-04 ± 
3.420e-05 

8.363e-04 ± 
4.308e-05 

8.002e-04 ± 
4.466e-05 

8.092e-04 ± 
4.166e-05 

8.408e-04 ± 
6.470e-05 

8.724e-04 ± 
7.740e-05 

8.542e-04 ± 
6.646e-05 

8.414e-04 ± 
6.471e-05 

8.427e-04 ± 
6.318e-05 

8.668e-04 ± 
9.904e-05 

8.868e-04 ± 
1.308e-04 

8.757e-04 ± 
1.424e-04 

8.500e-04 ± 
8.756e-05 

8.773e-04 ± 
9.609e-05 

8.577e-04 ± 
8.279e-05 

8.148e-04 ± 
9.633e-05 

7.973e-04 ± 
7.958e-05 

8.342e-04 ± 
9.824e-05 

8.154e-04 ± 
1.009e-04 

8.310e-04 ± 
9.193e-05 

8.298e-04 ± 
8.381e-05 

1.894e-03 ± 
4.438e-04 

1.808e-03 ± 
4.573e-04 

7.909e-04 ± 
9.539e-05 

7.661e-04 ± 
9.179e-05 

8.222e-04 ± 
5.424e-05 

8.221e-04 ± 
4.221e-05 

8.265e-04 ± 
6.365e-05 

8.295e-04 ± 
6.147e-05 

 305 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.88 % 

-1.19 % 

1.807 

0.583 

0.957 

0.994 

-0.52 % 

-1.04 % 

1.451 

0.612 

0.934 

0.994 

-0.78 % 

-2.56 % 

1.078 

0.612 

1.575 

0.994 

-1.26 % 

-2.16 % 

1.991 

0.583 

1.361 

0.994 

-0.5 % 

-1.88 % 

0.442 

0.794 

0.818 

0.994 

-0.78 % 

-3.44 % 

0.792 

0.753 

2.452 

0.994 

0 % 

-1.83 % 

-0.202 

0.824 

0.996 

0.994 

0.59 % 

-1.15 % 

-0.677 

0.753 

0.484 

0.994 

0.88 % 

0.89 % 

-1.171 

0.612 

-0.107 

0.994 

1.49 % 

0.69 % 

-1.890 

0.583 

-0.086 

0.994 

0.48 % 

0.85 % 

-1.228 

0.612 

-0.222 

0.994 

-0.01 % 

1.52 % 

-0.198 

0.824 

-0.371 

0.994 

0.24 % 

-0.23 % 

-0.793 

0.753 

0.242 

0.994 

-0.38 % 

0.52 % 

0.516 

0.762 

-0.113 

0.994 

-0.5 % 

-0.52 % 

0.647 

0.753 

0.250 

0.994 

-0.3 % 

-0.52 % 

0.429 

0.794 

0.252 

0.994 

-0.34 % 

1.54 % 

0.303 

0.812 

-0.478 

0.994 

-0.15 % 

-2.23 % 

-0.016 

0.834 

0.769 

0.994 

-0.14 % 

-2.37 % 

-0.013 

0.834 

1.015 

0.994 

-1.44 % 

-0.51 % 

1.110 

0.612 

0.348 

0.994 

-0.94 % 

-3.94 % 

0.350 

0.803 

1.471 

0.994 

-1.07 % 

-1.48 % 

0.710 

0.753 

0.431 

0.994 

-0.78 % 

-4.34 % 

0.528 

0.762 

1.496 

0.994 

-0.25 % 

-0.35 % 

0.400 

0.794 

0.373 

0.994 

-0.49 % 

-0.62 % 

0.718 

0.753 

1.461 

0.994 

-1.05 % 

-0.68 % 

1.879 

0.583 

0.558 

0.994 

-0.88 % 

-0.82 % 

1.405 

0.612 

1.734 

0.994 

 
 
 
 
 
 
 
 
Table A.22 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

7.709e-04 ± 
6.400e-05 

8.491e-04 ± 
3.718e-05 

8.383e-04 ± 
4.026e-05 

8.771e-04 ± 
5.618e-05 

8.644e-04 ± 
6.271e-05 

8.766e-04 ± 
3.370e-05 

8.538e-04 ± 
3.807e-05 

8.344e-04 ± 
3.701e-05 

8.222e-04 ± 
2.993e-05 

8.436e-04 ± 
3.142e-05 

8.262e-04 ± 
2.952e-05 

8.623e-04 ± 
3.992e-05 

8.517e-04 ± 
3.760e-05 

9.040e-04 ± 
5.172e-05 

8.819e-04 ± 
4.788e-05 

8.215e-04 ± 
2.976e-05 

7.966e-04 ± 
2.950e-05 

8.314e-04 ± 
4.660e-05 

8.349e-04 ± 
4.480e-05 

8.532e-04 ± 
4.461e-05 

8.356e-04 ± 
4.836e-05 

8.693e-04 ± 
3.837e-05 

8.433e-04 ± 
3.958e-05 

8.801e-04 ± 
3.850e-05 

8.407e-04 ± 
3.839e-05 

8.719e-04 ± 
3.638e-05 

8.407e-04 ± 
3.774e-05 

7.697e-04 ± 
4.772e-05 

8.474e-04 ± 
3.487e-05 

7.622e-04 ± 
6.892e-05 

7.848e-04 ± 
9.616e-05 

8.396e-04 ± 
4.156e-05 

8.338e-04 ± 
5.953e-05 

8.390e-04 ± 
3.575e-05 

8.361e-04 ± 
4.832e-05 

8.307e-04 ± 
6.177e-05 

8.731e-04 ± 
5.497e-05 

8.611e-04 ± 
5.522e-05 

8.809e-04 ± 
3.966e-05 

8.567e-04 ± 
3.692e-05 

8.281e-04 ± 
4.502e-05 

8.166e-04 ± 
3.131e-05 

8.413e-04 ± 
3.709e-05 

8.266e-04 ± 
2.818e-05 

8.568e-04 ± 
4.166e-05 

8.470e-04 ± 
3.496e-05 

9.069e-04 ± 
6.034e-05 

8.793e-04 ± 
5.270e-05 

8.593e-04 ± 
4.323e-05 

8.620e-04 ± 
9.727e-05 

8.667e-04 ± 
5.107e-05 

8.627e-04 ± 
7.361e-05 

8.763e-04 ± 
6.043e-05 

8.723e-04 ± 
1.045e-04 

8.475e-04 ± 
5.643e-05 

8.504e-04 ± 
9.830e-05 

8.059e-04 ± 
6.118e-05 

7.971e-04 ± 
8.812e-05 

8.140e-04 ± 
7.787e-05 

7.927e-04 ± 
5.724e-05 

8.230e-04 ± 
6.739e-05 

8.117e-04 ± 
7.424e-05 

8.238e-04 ± 
4.219e-05 

8.184e-04 ± 
6.068e-05 

8.461e-04 ± 
5.588e-05 

8.326e-04 ± 
7.016e-05 

8.409e-04 ± 
5.194e-05 

8.283e-04 ± 
5.991e-05 

9.201e-04 ± 
1.167e-04 

9.083e-04 ± 
1.389e-04 

8.841e-04 ± 
8.675e-05 

8.738e-04 ± 
1.198e-04 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.16 % 

2.97 % 

0.058 

0.834 

-0.685 

0.994 

-0.19 % 

-0.69 % 

0.119 

0.824 

0.502 

0.994 

0.09 % 

-0.65 % 

-0.732 

0.753 

0.462 

0.994 

-0.45 % 

0.31 % 

0.507 

0.762 

-0.095 

0.994 

-0.38 % 

-0.45 % 

0.033 

0.834 

0.941 

0.994 

0.5 % 

-0.45 % 

-1.223 

0.612 

0.247 

0.994 

0.34 % 

0.34 % 

-1.041 

0.612 

-0.137 

0.994 

-0.76 % 

-1.09 % 

1.052 

0.612 

0.315 

0.994 

-0.68 % 

-2.61 % 

1.340 

0.612 

1.180 

0.994 

-0.27 % 

-1.37 % 

0.554 

0.762 

0.578 

0.994 

0.04 % 

-0.65 % 

-0.167 

0.824 

0.346 

0.994 

-0.64 % 

-1.59 % 

1.368 

0.612 

0.971 

0.994 

-0.56 % 

-1.5 % 

1.080 

0.612 

1.070 

0.994 

0.33 % 

-1.29 % 

-0.506 

0.762 

0.433 

0.994 

-0.3 % 

-1.17 % 

0.367 

0.803 

0.490 

0.994 

8.158e-04 ± 
4.965e-05 

7.739e-04 ± 
5.671e-05 

7.817e-04 ± 
1.379e-04 

-0.69 % 

1 % 

0.926 

0.68 

-0.286 

0.994 

7.905e-04 ± 
3.606e-05 

8.305e-04 ± 
5.370e-05 

8.297e-04 ± 
4.323e-05 

8.493e-04 ± 
4.268e-05 

8.321e-04 ± 
4.202e-05 

8.707e-04 ± 
3.993e-05 

8.438e-04 ± 
3.587e-05 

8.776e-04 ± 
5.074e-05 

8.388e-04 ± 
4.098e-05 

8.652e-04 ± 
5.058e-05 

7.778e-04 ± 
6.669e-05 

7.659e-04 ± 
8.547e-05 

7.872e-04 ± 
4.490e-05 

7.849e-04 ± 
7.971e-05 

8.093e-04 ± 
5.486e-05 

7.900e-04 ± 
4.325e-05 

8.269e-04 ± 
2.504e-05 

8.290e-04 ± 
6.754e-05 

8.240e-04 ± 
2.973e-05 

8.235e-04 ± 
4.740e-05 

8.646e-04 ± 
5.316e-05 

8.610e-04 ± 
9.075e-05 

8.355e-04 ± 
4.359e-05 

8.366e-04 ± 
7.779e-05 

8.617e-04 ± 
7.188e-05 

8.503e-04 ± 
1.108e-04 

8.364e-04 ± 
6.896e-05 

8.402e-04 ± 
1.130e-04 

8.324e-04 ± 
6.067e-05 

8.307e-04 ± 
9.639e-05 

-0.77 % 

-1.53 % 

1.200 

0.612 

0.585 

0.994 

-0.11 % 

-0.29 % 

-0.113 

0.824 

0.132 

0.994 

-0.62 % 

-2.38 % 

0.962 

0.666 

1.986 

0.994 

-0.46 % 

0.26 % 

0.761 

0.753 

-0.042 

0.994 

-0.42 % 

-0.07 % 

0.192 

0.824 

1.316 

0.994 

0.16 % 

-0.41 % 

-0.643 

0.753 

0.323 

0.994 

0.06 % 

0.14 % 

-0.636 

0.753 

0.008 

0.994 

-0.29 % 

-1.32 % 

0.396 

0.794 

0.531 

0.994 

-0.22 % 

0.45 % 

0.238 

0.824 

-0.013 

0.994 

-0.77 % 

-0.2 % 

1.305 

0.612 

0.217 

0.994 

8.328e-04 ± 
3.523e-05 

8.217e-04 ± 
5.214e-05 

8.208e-04 ± 
8.085e-05 

-0.94 % 

-0.11 % 

1.369 

0.612 

0.243 

0.994 

 306 

 
 
 
 
 
 
 
Table A.22 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

8.546e-04 ± 
3.296e-05 

8.532e-04 ± 
3.367e-05 

8.681e-04 ± 
3.775e-05 

8.483e-04 ± 
3.205e-05 

8.854e-04 ± 
5.808e-05 

8.678e-04 ± 
6.321e-05 

8.839e-04 ± 
4.589e-05 

8.672e-04 ± 
5.007e-05 

8.748e-04 ± 
4.064e-05 

8.381e-04 ± 
4.468e-05 

8.735e-04 ± 
3.971e-05 

8.384e-04 ± 
4.199e-05 

8.774e-04 ± 
3.761e-05 

8.873e-04 ± 
4.435e-05 

8.901e-04 ± 
4.570e-05 

8.706e-04 ± 
4.872e-05 

8.323e-04 ± 
3.522e-05 

8.588e-04 ± 
3.190e-05 

8.492e-04 ± 
4.133e-05 

8.467e-04 ± 
3.379e-05 

8.606e-04 ± 
5.403e-05 

8.393e-04 ± 
3.708e-05 

8.808e-04 ± 
5.635e-05 

8.642e-04 ± 
5.531e-05 

8.860e-04 ± 
4.847e-05 

8.667e-04 ± 
4.562e-05 

8.759e-04 ± 
5.389e-05 

8.402e-04 ± 
4.898e-05 

8.202e-04 ± 
5.619e-05 

8.096e-04 ± 
6.461e-05 

8.282e-04 ± 
5.459e-05 

8.094e-04 ± 
4.894e-05 

8.250e-04 ± 
5.963e-05 

7.990e-04 ± 
5.342e-05 

8.203e-04 ± 
6.524e-05 

8.066e-04 ± 
6.358e-05 

8.676e-04 ± 
4.365e-05 

8.699e-04 ± 
9.549e-05 

8.694e-04 ± 
5.158e-05 

8.661e-04 ± 
7.456e-05 

8.881e-04 ± 
6.674e-05 

8.837e-04 ± 
1.144e-04 

8.681e-04 ± 
6.461e-05 

8.663e-04 ± 
1.052e-04 

8.789e-04 ± 
9.192e-05 

8.638e-04 ± 
1.311e-04 

8.543e-04 ± 
1.009e-04 

8.571e-04 ± 
1.489e-04 

8.671e-04 ± 
5.643e-05 

8.432e-04 ± 
7.421e-05 

8.399e-04 ± 
1.161e-04 

8.300e-04 ± 
3.974e-05 

8.712e-04 ± 
4.984e-05 

8.780e-04 ± 
4.565e-05 

8.816e-04 ± 
6.655e-05 

8.586e-04 ± 
5.193e-05 

8.361e-04 ± 
3.717e-05 

8.268e-04 ± 
6.507e-05 

8.240e-04 ± 
9.576e-05 

8.484e-04 ± 
7.055e-05 

8.334e-04 ± 
8.054e-05 

8.709e-04 ± 
7.717e-05 

8.488e-04 ± 
7.828e-05 

8.578e-04 ± 
7.861e-05 

8.281e-04 ± 
8.883e-05 

8.573e-04 ± 
1.056e-04 

8.376e-04 ± 
1.012e-04 

8.083e-04 ± 
4.128e-05 

8.006e-04 ± 
5.181e-05 

8.573e-04 ± 
3.760e-05 

8.412e-04 ± 
5.533e-05 

8.217e-04 ± 
5.704e-05 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-0.62 % 

-1.28 % 

1.260 

0.612 

0.698 

0.994 

-0.76 % 

-2.27 % 

1.988 

0.583 

1.737 

0.994 

-0.86 % 

-3.15 % 

1.395 

0.612 

1.973 

0.994 

-1.05 % 

-1.67 % 

2.629 

0.583 

1.139 

0.994 

-0.52 % 

0.27 % 

0.627 

0.753 

-0.071 

0.994 

-0.42 % 

-0.38 % 

0.127 

0.824 

0.954 

0.994 

0.25 % 

-0.49 % 

-0.601 

0.759 

0.282 

0.994 

-0.06 % 

-0.21 % 

-0.216 

0.824 

0.175 

0.994 

0.12 % 

-1.72 % 

-0.142 

0.824 

0.535 

0.994 

0.25 % 

0.32 % 

-0.330 

0.805 

0.031 

0.994 

-0.73 % 

-0.38 % 

1.143 

0.612 

0.220 

0.994 

-1 % 

-0.34 % 

1.221 

0.612 

0.268 

0.994 

-0.71 % 

-1.77 % 

1.098 

0.612 

0.782 

0.994 

-1.04 % 

-2.54 % 

1.803 

0.583 

1.388 

0.994 

-0.95 % 

-3.46 % 

1.107 

0.612 

1.424 

0.994 

-1.38 % 

-2.3 % 

1.843 

0.583 

0.987 

0.994 

0.46 % 

-0.95 % 

-1.161 

0.612 

0.498 

0.994 

-0.17 % 

-2.32 % 

0.096 

0.824 

1.021 

0.994 

Tables A.22- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 307 

 
 
 
 
 
 
 
 
 
Table A.23 Results of Post Hoc Tract Specific Comparison of Neurite Orientation Dispersion and 
Density Imaging Orientation Dispersion Index 

mTBI 

Controls 

P1 

P2 

P1 

P2 

0.351 ± 
0.015 

0.346 ± 
0.010 

0.320 ± 
0.013 

0.323 ± 
0.011 

0.283 ± 
0.018 

0.247 ± 
0.017 

0.303 ± 
0.012 

0.309 ± 
0.017 

0.299 ± 
0.015 

0.285 ± 
0.015 

0.322 ± 
0.015 

0.319 ± 
0.013 

0.312 ± 
0.012 

0.327 ± 
0.018 

0.325 ± 
0.021 

0.265 ± 
0.012 

0.270 ± 
0.011 

0.267 ± 
0.012 

0.265 ± 
0.011 

0.140 ± 
0.043 

0.135 ± 
0.041 

0.357 ± 
0.018 

0.372 ± 
0.017 

0.319 ± 
0.011 

0.323 ± 
0.009 

0.320 ± 
0.012 

0.354 ± 
0.012 

0.350 ± 
0.011 

0.322 ± 
0.012 

0.326 ± 
0.012 

0.286 ± 
0.022 

0.252 ± 
0.035 

0.305 ± 
0.012 

0.312 ± 
0.018 

0.302 ± 
0.017 

0.290 ± 
0.016 

0.324 ± 
0.015 

0.322 ± 
0.016 

0.315 ± 
0.012 

0.326 ± 
0.020 

0.326 ± 
0.017 

0.266 ± 
0.012 

0.272 ± 
0.012 

0.268 ± 
0.012 

0.267 ± 
0.012 

0.147 ± 
0.048 

0.142 ± 
0.043 

0.359 ± 
0.018 

0.376 ± 
0.018 

0.321 ± 
0.013 

0.326 ± 
0.012 

0.324 ± 
0.021 

0.362 ± 
0.016 

0.356 ± 
0.014 

0.327 ± 
0.012 

0.332 ± 
0.012 

0.286 ± 
0.016 

0.248 ± 
0.022 

0.313 ± 
0.016 

0.319 ± 
0.026 

0.310 ± 
0.019 

0.299 ± 
0.020 

0.333 ± 
0.019 

0.325 ± 
0.015 

0.322 ± 
0.015 

0.337 ± 
0.022 

0.337 ± 
0.019 

0.277 ± 
0.014 

0.279 ± 
0.013 

0.281 ± 
0.017 

0.279 ± 
0.013 

0.154 ± 
0.062 

0.152 ± 
0.058 

0.376 ± 
0.023 

0.390 ± 
0.018 

0.326 ± 
0.013 

0.329 ± 
0.014 

0.325 ± 
0.011 

0.365 ± 
0.017 

0.362 ± 
0.020 

0.331 ± 
0.015 

0.337 ± 
0.018 

0.301 ± 
0.042 

0.263 ± 
0.046 

0.319 ± 
0.018 

0.327 ± 
0.026 

0.317 ± 
0.023 

0.305 ± 
0.025 

0.337 ± 
0.021 

0.333 ± 
0.033 

0.328 ± 
0.019 

0.344 ± 
0.026 

0.345 ± 
0.024 

0.281 ± 
0.015 

0.287 ± 
0.020 

0.287 ± 
0.018 

0.283 ± 
0.016 

0.178 ± 
0.079 

0.167 ± 
0.079 

0.374 ± 
0.026 

0.385 ± 
0.023 

0.335 ± 
0.028 

0.334 ± 
0.025 

0.339 ± 
0.037 

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

0.9 % 

0.86 % 

-1.495 

0.008* 

-0.520 

0.127 

1.19 % 

1.71 % 

-2.701 

0.005* 

-0.970 

0.114 

0.58 % 

1.29 % 

-1.351 

0.008* 

-1.084 

0.114 

0.82 % 

1.47 % 

-1.908 

0.007* 

-1.139 

0.114 

1.28 % 

5.3 % 

-1.216 

0.008* 

-1.406 

0.114 

1.96 % 

5.9 % 

-1.313 

0.008* 

-0.966 

0.114 

0.76 % 

2.09 % 

-1.469 

0.008* 

-1.007 

0.114 

1.04 % 

2.49 % 

-1.217 

0.008* 

-0.814 

0.114 

1.16 % 

2.33 % 

-1.647 

0.007* 

-1.058 

0.114 

1.63 % 

1.84 % 

-2.083 

0.007* 

-0.679 

0.121 

0.62 % 

1.22 % 

-0.849 

0.012* 

-0.521 

0.127 

0.65 % 

2.34 % 

-1.102 

0.01* 

-0.878 

0.114 

0.73 % 

2.03 % 

-1.401 

0.008* 

-0.970 

0.114 

-0.28 % 

2.16 % 

0.128 

0.022* 

-0.890 

0.114 

0.36 % 

2.42 % 

-0.247 

0.02* 

-0.691 

0.121 

0.52 % 

1.67 % 

-0.948 

0.011* 

-1.075 

0.114 

0.82 % 

2.65 % 

-1.545 

0.007* 

-1.487 

0.114 

0.32 % 

1.94 % 

-0.889 

0.011* 

-0.927 

0.114 

0.62 % 

1.43 % 

-1.411 

0.008* 

-0.850 

0.114 

5.17 % 

15.52 % 

-2.011 

0.007* 

-1.518 

0.114 

5.24 % 

10.19 % 

-2.069 

0.007* 

-0.797 

0.114 

0.82 % 

-0.62 % 

-1.784 

0.007* 

0.553 

0.127 

1.28 % 

-1.51 % 

-2.828 

0.005* 

0.963 

0.114 

0.67 % 

2.96 % 

-1.308 

0.008* 

-1.078 

0.114 

0.85 % 

1.63 % 

-1.883 

0.007* 

-0.831 

0.114 

1.36 % 

4.52 % 

-1.584 

0.007* 

-1.442 

0.114 

 308 

 
 
 
 
 
 
 
 
Table A.23 (cont’d) 

mTBI 

Controls 

TractSeg 
Name 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

P1 

0.324 ± 
0.015 

0.344 ± 
0.015 

0.348 ± 
0.015 

0.344 ± 
0.015 

0.309 ± 
0.013 

0.305 ± 
0.012 

0.294 ± 
0.014 

0.289 ± 
0.013 

0.300 ± 
0.013 

0.303 ± 
0.010 

0.340 ± 
0.017 

0.337 ± 
0.012 

0.363 ± 
0.016 

0.364 ± 
0.015 

0.360 ± 
0.017 

0.352 ± 
0.016 

0.270 ± 
0.012 

0.272 ± 
0.014 

0.295 ± 
0.017 

0.287 ± 
0.016 

0.312 ± 
0.013 

P2 

P1 

0.328 ± 0.022  0.326 ± 0.018 

0.347 ± 0.015  0.359 ± 0.020 

0.349 ± 0.014  0.357 ± 0.017 

0.347 ± 0.015  0.354 ± 0.018 

0.311 ± 0.014  0.316 ± 0.012 

0.306 ± 0.013  0.309 ± 0.018 

0.296 ± 0.013  0.306 ± 0.015 

0.291 ± 0.014  0.297 ± 0.015 

0.303 ± 0.014  0.319 ± 0.016 

0.307 ± 0.014  0.320 ± 0.016 

0.344 ± 0.015  0.353 ± 0.024 

0.343 ± 0.017  0.352 ± 0.019 

0.362 ± 0.016  0.372 ± 0.017 

0.365 ± 0.015  0.376 ± 0.017 

0.360 ± 0.017  0.371 ± 0.016 

0.353 ± 0.018  0.362 ± 0.016 

0.272 ± 0.012  0.283 ± 0.013 

0.275 ± 0.013  0.283 ± 0.014 

0.299 ± 0.025  0.301 ± 0.018 

0.290 ± 0.026  0.294 ± 0.017 

0.313 ± 0.015  0.321 ± 0.014 

P2 

0.336 ± 
0.032 

0.359 ± 
0.023 

0.359 ± 
0.017 

0.359 ± 
0.020 

0.326 ± 
0.032 

0.316 ± 
0.029 

0.308 ± 
0.015 

0.301 ± 
0.016 

0.319 ± 
0.021 

0.319 ± 
0.020 

0.356 ± 
0.018 

0.360 ± 
0.029 

0.371 ± 
0.027 

0.379 ± 
0.017 

0.377 ± 
0.017 

0.364 ± 
0.018 

0.292 ± 
0.025 

0.287 ± 
0.016 

0.309 ± 
0.035 

0.302 ± 
0.028 

0.330 ± 
0.035 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

1.07 % 

3.19 % 

-1.415 

0.008* 

-1.157 

0.114 

0.87 % 

0.22 % 

-2.284 

0.007* 

-0.004 

0.191 

0.25 % 

0.58 % 

-0.230 

0.02* 

-0.397 

0.143 

0.91 % 

1.42 % 

-1.623 

0.007* 

-0.581 

0.127 

0.56 % 

3.12 % 

-0.898 

0.011* 

-1.214 

0.114 

0.38 % 

2.38 % 

-0.691 

0.014* 

-0.956 

0.114 

0.5 % 

0.73 % 

-0.814 

0.012* 

-0.572 

0.127 

0.67 % 

1.33 % 

-1.172 

0.009* 

-0.931 

0.114 

0.86 % 

0.06 % 

-1.932 

0.007* 

-0.012 

0.191 

1.3 % 

-0.5 % 

-2.530 

0.006* 

0.158 

0.172 

1.33 % 

0.91 % 

-1.916 

0.007* 

-0.381 

0.144 

1.87 % 

2.05 % 

-2.966 

0.005* 

-0.784 

0.114 

-0.26 % 

-0.3 % 

0.330 

0.019* 

0.234 

0.163 

0.35 % 

0.7 % 

-0.860 

0.012* 

-0.556 

0.127 

0 % 

1.51 % 

0.070 

0.023* 

-1.096 

0.114 

0.12 % 

0.66 % 

-0.346 

0.019* 

-0.556 

0.127 

0.76 % 

3.44 % 

-1.690 

0.007* 

-1.643 

0.114 

1.04 % 

1.62 % 

-2.047 

0.007* 

-1.155 

0.114 

1.52 % 

2.47 % 

-1.598 

0.007* 

-0.985 

0.114 

1.19 % 

2.81 % 

-1.212 

0.008* 

-0.715 

0.121 

0.47 % 

2.78 % 

-0.710 

0.014* 

-0.973 

0.114 

 309 

 
 
 
 
 
 
 
Table A.23 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

0.315 ± 
0.013 

0.327 ± 
0.014 

0.319 ± 
0.013 

0.305 ± 
0.015 

0.301 ± 
0.014 

0.306 ± 
0.013 

0.305 ± 
0.010 

0.312 ± 
0.012 

0.312 ± 
0.010 

0.325 ± 
0.015 

0.318 ± 
0.014 

0.313 ± 
0.012 

0.311 ± 
0.012 

0.326 ± 
0.013 

0.316 ± 
0.013 

0.304 ± 
0.014 

0.301 ± 
0.015 

0.303 ± 
0.012 

0.308 ± 
0.011 

0.307 ± 
0.011 

0.314 ± 
0.010 

0.312 ± 
0.013 

0.309 ± 
0.013 

0.300 ± 
0.014 

0.317 ± 
0.014 

0.328 ± 
0.014 

0.322 ± 
0.013 

0.308 ± 
0.014 

0.305 ± 
0.014 

0.309 ± 
0.011 

0.308 ± 
0.012 

0.313 ± 
0.011 

0.314 ± 
0.012 

0.327 ± 
0.014 

0.320 ± 
0.014 

0.314 ± 
0.014 

0.312 ± 
0.013 

0.326 ± 
0.013 

0.317 ± 
0.014 

0.307 ± 
0.014 

0.304 ± 
0.016 

0.305 ± 
0.012 

0.310 ± 
0.012 

0.308 ± 
0.010 

0.315 ± 
0.011 

0.314 ± 
0.012 

0.310 ± 
0.013 

0.303 ± 
0.022 

0.321 ± 
0.020 

0.337 ± 
0.015 

0.327 ± 
0.014 

0.319 ± 
0.012 

0.310 ± 
0.013 

0.318 ± 
0.011 

0.314 ± 
0.012 

0.322 ± 
0.015 

0.323 ± 
0.012 

0.339 ± 
0.016 

0.330 ± 
0.014 

0.319 ± 
0.011 

0.314 ± 
0.017 

0.335 ± 
0.013 

0.323 ± 
0.012 

0.319 ± 
0.015 

0.311 ± 
0.018 

0.314 ± 
0.012 

0.316 ± 
0.011 

0.318 ± 
0.014 

0.324 ± 
0.012 

0.324 ± 
0.012 

0.318 ± 
0.012 

0.307 ± 
0.017 

0.326 ± 
0.026 

0.341 ± 
0.015 

0.331 ± 
0.016 

0.324 ± 
0.013 

0.315 ± 
0.018 

0.326 ± 
0.013 

0.320 ± 
0.015 

0.330 ± 
0.017 

0.328 ± 
0.016 

0.348 ± 
0.016 

0.338 ± 
0.023 

0.329 ± 
0.031 

0.322 ± 
0.028 

0.338 ± 
0.016 

0.327 ± 
0.014 

0.322 ± 
0.016 

0.316 ± 
0.020 

0.321 ± 
0.015 

0.322 ± 
0.014 

0.324 ± 
0.017 

0.329 ± 
0.014 

0.330 ± 
0.015 

0.325 ± 
0.018 

0.319 ± 
0.035 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

0.65 % 

1.83 % 

-1.007 

0.011* 

-0.798 

0.114 

0.21 % 

0.98 % 

-0.341 

0.019* 

-0.690 

0.121 

0.86 % 

1.18 % 

-1.672 

0.007* 

-0.806 

0.114 

0.87 % 

1.51 % 

-1.554 

0.007* 

-0.987 

0.114 

1.29 % 

1.65 % 

-2.271 

0.007* 

-1.322 

0.114 

0.78 % 

2.53 % 

-1.501 

0.008* 

-2.009 

0.114 

0.98 % 

1.94 % 

-2.104 

0.007* 

-1.782 

0.114 

0.41 % 

2.42 % 

-1.192 

0.009* 

-1.380 

0.114 

0.76 % 

1.69 % 

-2.116 

0.007* 

-1.274 

0.114 

0.57 % 

2.53 % 

-1.544 

0.007* 

-1.637 

0.114 

0.64 % 

2.42 % 

-1.243 

0.008* 

-1.293 

0.114 

0.56 % 

3.07 % 

-0.940 

0.011* 

-1.228 

0.114 

0.34 % 

2.31 % 

-0.621 

0.015* 

-0.984 

0.114 

0.22 % 

1.02 % 

-0.417 

0.018* 

-0.779 

0.114 

0.51 % 

1.19 % 

-1.029 

0.01* 

-0.834 

0.114 

0.95 % 

0.84 % 

-1.667 

0.007* 

-0.664 

0.121 

1.13 % 

1.64 % 

-1.696 

0.007* 

-0.953 

0.114 

0.65 % 

2.32 % 

-1.348 

0.008* 

-1.962 

0.114 

0.6 % 

2.01 % 

-1.334 

0.008* 

-1.932 

0.114 

0.32 % 

2.13 % 

-0.992 

0.011* 

-1.365 

0.114 

0.56 % 

1.56 % 

-1.552 

0.007* 

-1.465 

0.114 

0.54 % 

1.82 % 

-1.417 

0.008* 

-1.442 

0.114 

0.28 % 

2.21 % 

-0.471 

0.017* 

-1.244 

0.114 

1.2 % 

3.62 % 

-1.373 

0.008* 

-1.163 

0.114 

UF_right 

0.297 ± 
0.014 

0.301 ± 
0.021 

0.315 ± 
0.032 
Tables A.23- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

0.305 ± 
0.015 

3.28 % 

1.49 % 

0.007* 

-1.906 

-1.016 

0.114 

 310 

 
 
 
 
 
 
 
 
Table A.24 Results of Post Hoc Tract Specific Comparison of Neurite Orientation Dispersion and 
Density Imaging Neurite Density Index 

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-
score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

AF_left 

AF_right 

ATR_left 

ATR_right 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

0.482 ± 0.022 

0.481 ± 0.021 

0.493 ± 0.030 

0.497 ± 0.025 

-0.25 % 

0.75 % 

0.768 

0.532 

-0.925 

0.503 ± 0.023 

0.501 ± 0.019 

0.507 ± 0.032 

0.508 ± 0.034 

-0.37 % 

0.12 % 

1.266 

0.33 

-0.052 

0.478 ± 0.019 

0.479 ± 0.020 

0.499 ± 0.029 

0.503 ± 0.024 

0.13 % 

0.73 % 

-0.452 

0.603 

-0.910 

0.480 ± 0.019 

0.480 ± 0.018 

0.495 ± 0.023 

0.500 ± 0.025 

0.04 % 

1.03 % 

-0.009 

0.674 

-1.316 

0.401 ± 0.026 

0.396 ± 0.026 

0.426 ± 0.028 

0.445 ± 0.024 

-1.18 % 

4.44 % 

2.135 

0.13 

-2.270 

0.485 ± 0.031 

0.476 ± 0.037 

0.506 ± 0.031 

0.496 ± 0.050 

-1.73 % 

-1.84 % 

3.338 

0.037* 

0.920 

0.490 ± 0.023 

0.488 ± 0.023 

0.502 ± 0.033 

0.503 ± 0.029 

-0.36 % 

0.09 % 

1.166 

0.338 

-0.366 

0.493 ± 0.023 

0.494 ± 0.024 

0.509 ± 0.045 

0.511 ± 0.034 

0.1 % 

0.36 % 

-0.191 

0.659 

-0.478 

0.499 

0.588 

0.499 

0.499 

0.499 

0.499 

0.56 

0.56 

0.543 ± 0.026 

0.544 ± 0.029 

0.548 ± 0.041 

0.554 ± 0.038 

0.03 % 

1.05 % 

-0.045 

0.674 

-1.141 

0.499 

0.538 ± 0.024 

0.537 ± 0.025 

0.546 ± 0.036 

0.547 ± 0.035 

-0.3 % 

0.27 % 

0.778 

0.532 

-0.528 

0.497 ± 0.021 

0.494 ± 0.020 

0.498 ± 0.027 

0.497 ± 0.027 

-0.62 % 

-0.16 % 

2.070 

0.132 

0.514 ± 0.023 

0.512 ± 0.021 

0.516 ± 0.023 

0.514 ± 0.033 

-0.35 % 

-0.27 % 

1.182 

0.338 

0.079 

0.317 

0.501 ± 0.020 

0.499 ± 0.020 

0.508 ± 0.030 

0.508 ± 0.028 

-0.38 % 

0.08 % 

1.480 

0.276 

-0.327 

0.473 ± 0.025 

0.473 ± 0.025 

0.484 ± 0.033 

0.482 ± 0.035 

-0.1 % 

-0.58 % 

0.488 

0.602 

0.338 

0.482 ± 0.025 

0.480 ± 0.022 

0.488 ± 0.033 

0.493 ± 0.030 

-0.46 % 

1.13 % 

1.011 

0.408 

-1.090 

0.598 ± 0.021 

0.597 ± 0.023 

0.607 ± 0.034 

0.619 ± 0.034 

-0.03 % 

1.96 % 

0.137 

0.674 

-1.876 

0.600 ± 0.022 

0.599 ± 0.023 

0.607 ± 0.033 

0.610 ± 0.042 

-0.18 % 

0.48 % 

0.871 

0.486 

-0.370 

0.559 ± 0.022 

0.559 ± 0.024 

0.576 ± 0.038 

0.580 ± 0.042 

0.06 % 

0.62 % 

0.037 

0.674 

-0.266 

0.559 ± 0.020 

0.558 ± 0.022 

0.572 ± 0.034 

0.579 ± 0.033 

-0.21 % 

1.12 % 

0.731 

0.532 

-1.064 

0.451 ± 0.053 

0.451 ± 0.048 

0.463 ± 0.039 

0.481 ± 0.052 

-0.13 % 

3.88 % 

0.059 

0.674 

-1.174 

0.447 ± 0.053 

0.454 ± 0.056 

0.467 ± 0.042 

0.471 ± 0.044 

1.52 % 

0.69 % 

-1.339 

0.303 

0.218 

0.591 ± 0.028 

0.589 ± 0.036 

0.588 ± 0.042 

0.594 ± 0.060 

-0.36 % 

0.92 % 

0.733 

0.532 

-0.099 

0.577 ± 0.021 

0.573 ± 0.026 

0.571 ± 0.046 

0.584 ± 0.048 

-0.64 % 

2.26 % 

1.432 

0.276 

-0.658 

0.482 ± 0.022 

0.477 ± 0.020 

0.491 ± 0.022 

0.491 ± 0.032 

-1.04 % 

-0.01 % 

3.550 

0.037* 

-0.057 

0.486 ± 0.022 

0.483 ± 0.020 

0.493 ± 0.021 

0.497 ± 0.035 

-0.69 % 

0.76 % 

2.500 

0.121 

-0.431 

0.467 ± 0.022 

0.463 ± 0.021 

0.478 ± 0.023 

0.483 ± 0.029 

-0.81 % 

1.02 % 

2.453 

0.121 

-1.322 

0.471 ± 0.021 

0.469 ± 0.020 

0.480 ± 0.026 

0.486 ± 0.033 

-0.51 % 

1.25 % 

1.564 

0.251 

-1.288 

0.617 ± 0.023 

0.615 ± 0.027 

0.610 ± 0.046 

0.622 ± 0.046 

-0.44 % 

1.91 % 

1.215 

0.33 

-0.818 

0.470 ± 0.021 

0.467 ± 0.020 

0.474 ± 0.024 

0.477 ± 0.023 

-0.56 % 

0.79 % 

2.286 

0.128 

-1.012 

0.477 ± 0.020 

0.474 ± 0.018 

0.477 ± 0.028 

0.479 ± 0.030 

-0.65 % 

0.31 % 

2.322 

0.128 

-0.291 

0.487 ± 0.022 

0.483 ± 0.022 

0.497 ± 0.020 

0.501 ± 0.033 

-0.77 % 

0.86 % 

2.041 

0.132 

-0.632 

0.493 ± 0.022 

0.490 ± 0.022 

0.499 ± 0.026 

0.505 ± 0.033 

-0.58 % 

1.2 % 

1.907 

0.159 

-1.314 

0.531 ± 0.019 

0.528 ± 0.019 

0.537 ± 0.026 

0.541 ± 0.028 

-0.67 % 

0.88 % 

2.226 

0.129 

-0.757 

0.538 ± 0.017 

0.537 ± 0.019 

0.545 ± 0.025 

0.548 ± 0.032 

-0.27 % 

0.55 % 

1.146 

0.34 

-0.210 

0.575 ± 0.021 

0.575 ± 0.028 

0.589 ± 0.039 

0.601 ± 0.046 

-0.01 % 

2.01 % 

0.077 

0.674 

-0.855 

0.572 ± 0.016 

0.570 ± 0.022 

0.577 ± 0.041 

0.591 ± 0.039 

-0.26 % 

2.32 % 

0.652 

0.565 

-1.228 

0.513 ± 0.026 

0.508 ± 0.026 

0.518 ± 0.040 

0.518 ± 0.034 

-1.03 % 

0.04 % 

2.069 

0.132 

-0.252 

0.517 ± 0.025 

0.513 ± 0.024 

0.519 ± 0.036 

0.514 ± 0.048 

-0.81 % 

-0.99 % 

1.708 

0.207 

0.487 

0.488 ± 0.023 

0.490 ± 0.026 

0.490 ± 0.039 

0.498 ± 0.033 

0.31 % 

1.57 % 

-0.491 

0.602 

-1.073 

0.488 ± 0.023 

0.488 ± 0.024 

0.488 ± 0.038 

0.495 ± 0.026 

0.02 % 

1.28 % 

0.204 

0.659 

-0.992 

0.464 ± 0.022 

0.465 ± 0.027 

0.466 ± 0.036 

0.472 ± 0.029 

0.22 % 

1.25 % 

-0.323 

0.629 

-1.038 

0.481 ± 0.023 

0.483 ± 0.026 

0.485 ± 0.038 

0.492 ± 0.028 

0.39 % 

1.46 % 

-0.543 

0.601 

-0.910 

0.590 ± 0.025 

0.589 ± 0.025 

0.605 ± 0.028 

0.609 ± 0.049 

-0.06 % 

0.57 % 

0.309 

0.629 

-0.214 

0.598 ± 0.021 

0.593 ± 0.023 

0.607 ± 0.025 

0.613 ± 0.030 

-0.74 % 

0.98 % 

2.204 

0.129 

-1.507 

0.465 ± 0.026 

0.462 ± 0.026 

0.492 ± 0.029 

0.496 ± 0.030 

-0.7 % 

0.93 % 

1.442 

0.276 

-0.659 

0.467 ± 0.027 

0.464 ± 0.029 

0.486 ± 0.023 

0.486 ± 0.046 

-0.64 % 

-0.02 % 

1.576 

0.251 

0.030 

0.56 

0.588 

0.56 

0.56 

0.56 

0.499 

0.499 

0.56 

0.56 

0.499 

0.499 

0.56 

0.588 

0.553 

0.588 

0.56 

0.499 

0.499 

0.516 

0.499 

0.56 

0.557 

0.499 

0.528 

0.56 

0.506 

0.499 

0.56 

0.56 

0.499 

0.499 

0.499 

0.499 

0.56 

0.499 

0.553 

0.59 

0.485 ± 0.022 

0.481 ± 0.021 

0.493 ± 0.020 

0.497 ± 0.033 

-0.89 % 

0.78 % 

2.885 

0.068 

-0.662 

0.553 

 311 

 
 
 
 
 
 
 
 
Table A.24 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

q-
value 
mTBI
1 

t-
score 
mTBI 

t-score 
control
s 

q-value 
control
s 

0.499 

0.499 

0.588 

0.499 

0.588 

0.499 

0.56 

0.56 

0.499 

0.499 

0.56 

0.56 

ST_OCC_right 

0.492 ± 0.022 

0.489 ± 0.021 

0.497 ± 0.023 

0.503 ± 0.032 

-0.65 % 

1.24 % 

2.144 

0.13 

-1.231 

ST_PAR_left 

0.494 ± 0.020 

0.491 ± 0.019 

0.501 ± 0.023 

0.504 ± 0.023 

-0.61 % 

0.67 % 

2.310  0.128 

-0.931 

ST_PAR_right 

0.509 ± 0.019 

0.508 ± 0.018 

0.515 ± 0.024 

0.517 ± 0.029 

-0.33 % 

0.28 % 

1.471  0.276 

-0.082 

ST_POSTC_left 

0.529 ± 0.019 

0.529 ± 0.021 

0.547 ± 0.023 

0.553 ± 0.023 

-0.14 % 

1.04 % 

0.418  0.603 

-1.128 

ST_POSTC_right 

0.546 ± 0.021 

0.547 ± 0.019 

0.560 ± 0.025 

0.562 ± 0.033 

0.18 % 

0.3 % 

-0.015  0.674 

-0.101 

ST_PREC_left 

0.547 ± 0.021 

0.547 ± 0.023 

0.563 ± 0.028 

0.571 ± 0.029 

0.04 % 

1.52 % 

-0.125  0.674 

-1.860 

ST_PREC_right 

0.558 ± 0.024 

0.558 ± 0.022 

0.571 ± 0.029 

0.573 ± 0.033 

-0.04 % 

0.41 % 

0.543  0.601 

-0.347 

ST_PREF_left 

0.503 ± 0.022 

0.502 ± 0.021 

0.521 ± 0.031 

0.522 ± 0.031 

-0.1 % 

0.09 % 

0.432  0.603 

-0.218 

ST_PREF_right 

0.504 ± 0.023 

0.503 ± 0.020 

0.519 ± 0.026 

0.524 ± 0.028 

-0.28 % 

0.91 % 

1.222 

0.33 

-1.166 

ST_PREM_left 

0.502 ± 0.021 

0.503 ± 0.021 

0.521 ± 0.033 

0.527 ± 0.027 

0.25 % 

1.2 % 

-0.647  0.565 

-1.478 

ST_PREM_right 

0.522 ± 0.023 

0.522 ± 0.021 

0.542 ± 0.034 

0.540 ± 0.033 

0.03 % 

-0.3 % 

-0.063  0.674 

0.281 

T_OCC_left 

0.486 ± 0.022 

0.482 ± 0.022 

0.496 ± 0.020 

0.499 ± 0.032 

-0.71 % 

0.7 % 

1.960 

T_OCC_right 

0.492 ± 0.021 

0.489 ± 0.021 

0.498 ± 0.025 

0.504 ± 0.033 

-0.52 % 

1.11 % 

1.796 

0.15 

0.19 

-0.526 

-1.238 

0.499 

T_PAR_left 

0.497 ± 0.019 

0.494 ± 0.020 

0.503 ± 0.025 

0.506 ± 0.025 

-0.5 % 

0.63 % 

1.713  0.207 

-0.790 

T_PAR_right 

0.510 ± 0.018 

0.508 ± 0.019 

0.516 ± 0.023 

0.517 ± 0.027 

-0.35 % 

0.35 % 

1.386  0.289 

-0.248 

0.52 

0.56 

T_POSTC_left 

0.529 ± 0.020 

0.528 ± 0.022 

0.541 ± 0.027 

0.548 ± 0.026 

-0.24 % 

1.23 % 

0.608 

0.58 

-1.117 

0.499 

T_POSTC_right 

0.539 ± 0.020 

0.539 ± 0.021 

0.546 ± 0.027 

0.549 ± 0.034 

0 % 

0.47 % 

0.211  0.659 

-0.514 

0.56 

T_PREC_left 

0.555 ± 0.021 

0.556 ± 0.024 

0.569 ± 0.031 

0.579 ± 0.031 

0.2 % 

1.85 % 

-0.465  0.603 

-2.211 

T_PREC_right 

0.559 ± 0.022 

0.559 ± 0.023 

0.568 ± 0.030 

0.571 ± 0.033 

-0.03 % 

0.61 % 

0.335  0.629 

-0.873 

0.499 

0.506 

T_PREF_left 

0.508 ± 0.021 

0.509 ± 0.022 

0.527 ± 0.032 

0.528 ± 0.032 

0.17 % 

0.21 % 

-0.321  0.629 

-0.240 

0.56 

T_PREF_right 

0.503 ± 0.020 

0.503 ± 0.020 

0.517 ± 0.027 

0.523 ± 0.025 

-0.1 % 

1.27 % 

0.504  0.602 

-1.776 

T_PREM_left 

0.507 ± 0.020 

0.510 ± 0.022 

0.529 ± 0.034 

0.537 ± 0.027 

0.53 % 

1.61 % 

-1.224 

0.33 

-1.856 

0.499 

0.499 

T_PREM_right 

0.521 ± 0.019 

0.521 ± 0.022 

0.539 ± 0.033 

0.541 ± 0.028 

0.07 % 

0.37 % 

-0.202  0.659 

-0.533 

0.56 

UF_left 

0.438 ± 0.026 

0.432 ± 0.026 

0.452 ± 0.025 

0.458 ± 0.019 

-1.24 % 

1.28 % 

2.963  0.068 

-1.504 

0.499 

0.430 ± 0.024 

0.426 ± 0.023 

UF_right 
Tables A.24- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

0.451 ± 0.026 

0.443 ± 0.018 

2.680  0.094 

-0.96 % 

1.74 % 

-1.334 

0.499 

 312 

 
 
 
 
 
 
 
 
 
 
 
Table A.25 Results of Post Hoc Tract Specific Comparison of Neurite Orientation Dispersion and 
Density Imaging Free Water Fraction 

mTBI 

Controls 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

TractSeg 
Name 

AF_left 

AF_right 

ATR_left 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

0.071 ± 0.012  0.066 ± 0.015  0.058 ± 0.009  0.055 ± 0.014 

-7.44 % 

-5.23 % 

0.076 ± 0.013  0.072 ± 0.013  0.069 ± 0.013  0.065 ± 0.012 

-4.97 % 

-6.23 % 

0.090 ± 0.016  0.086 ± 0.022  0.082 ± 0.018  0.072 ± 0.018 

-4.3 % 

-11.69 % 

ATR_right 

0.097 ± 0.019  0.091 ± 0.019  0.092 ± 0.024  0.084 ± 0.025 

-5.92 % 

-8.32 % 

0.069 ± 0.026  0.061 ± 0.025  0.054 ± 0.021  0.053 ± 0.029 

-10.8 % 

-0.49 % 

0.100 ± 0.026  0.092 ± 0.031  0.106 ± 0.032  0.087 ± 0.023 

-7.51 % 

-18.55 % 

0.099 ± 0.013  0.097 ± 0.020  0.092 ± 0.015  0.082 ± 0.018 

-2.58 % 

-10.93 % 

0.114 ± 0.021  0.113 ± 0.031  0.103 ± 0.026  0.095 ± 0.032 

-0.46 % 

-8.03 % 

0.141 ± 0.019  0.142 ± 0.028  0.130 ± 0.023  0.127 ± 0.041 

0.37 % 

0.131 ± 0.019  0.133 ± 0.026  0.126 ± 0.027  0.121 ± 0.044 

1.86 % 

-2.04 % 

-3.46 % 

-0.193 

-0.876 

0.081 ± 0.017  0.080 ± 0.015  0.078 ± 0.012  0.077 ± 0.025 

-0.65 % 

-0.53 % 

0.094 ± 0.027  0.093 ± 0.019  0.093 ± 0.019  0.095 ± 0.029 

-1.28 % 

2.33 % 

0.097 ± 0.014  0.096 ± 0.016  0.093 ± 0.013  0.088 ± 0.023 

-1.29 % 

-5.23 % 

CG_left 

0.060 ± 0.012  0.059 ± 0.014  0.057 ± 0.014  0.053 ± 0.021 

-2.33 % 

-6.26 % 

CG_right 

0.060 ± 0.011  0.056 ± 0.012  0.054 ± 0.009  0.053 ± 0.025 

-6.13 % 

-2.81 % 

CST_left 

0.168 ± 0.017  0.164 ± 0.025  0.143 ± 0.016  0.139 ± 0.027 

-2.3 % 

-2.3 % 

CST_right 

0.145 ± 0.015  0.142 ± 0.017  0.131 ± 0.016  0.130 ± 0.025 

-2.41 % 

-0.92 % 

FPT_left 

0.144 ± 0.015  0.141 ± 0.022  0.128 ± 0.019  0.118 ± 0.021 

-1.77 % 

-7.88 % 

FPT_right 

0.140 ± 0.013  0.136 ± 0.017  0.128 ± 0.019  0.120 ± 0.022 

-2.35 % 

-6.47 % 

FX_left 

FX_right 

ICP_left 

0.501 ± 0.108  0.490 ± 0.103  0.485 ± 0.103  0.476 ± 0.123 

-2.24 % 

-2 % 

0.491 ± 0.104  0.483 ± 0.108  0.482 ± 0.104  0.451 ± 0.136 

-1.76 % 

-6.31 % 

0.095 ± 0.029  0.089 ± 0.016  0.089 ± 0.013  0.092 ± 0.018 

-5.86 % 

2.98 % 

ICP_right 

0.082 ± 0.026  0.077 ± 0.017  0.081 ± 0.019  0.081 ± 0.025 

-5.93 % 

-0.41 % 

IFO_left 

0.074 ± 0.014  0.069 ± 0.015  0.068 ± 0.012  0.065 ± 0.014 

-6.55 % 

-4.94 % 

IFO_right 

0.072 ± 0.016  0.067 ± 0.014  0.069 ± 0.014  0.067 ± 0.020 

-5.81 % 

-2.47 % 

ILF_left 

0.070 ± 0.020  0.063 ± 0.018  0.063 ± 0.015  0.061 ± 0.019 

-10.73 % 

-2.8 % 

ILF_right 

0.067 ± 0.023  0.061 ± 0.019  0.064 ± 0.018  0.065 ± 0.026 

-8.04 % 

2.2 % 

MCP 

0.095 ± 0.027  0.092 ± 0.019  0.085 ± 0.012  0.097 ± 0.026 

-2.89 % 

14.42 % 

MLF_left 

0.064 ± 0.015  0.061 ± 0.014  0.061 ± 0.012  0.060 ± 0.020 

-4.84 % 

-2.67 % 

MLF_right 

0.062 ± 0.017  0.060 ± 0.015  0.061 ± 0.013  0.059 ± 0.019 

-2.78 % 

-2.82 % 

OR_left 

0.090 ± 0.024  0.085 ± 0.022  0.086 ± 0.016  0.087 ± 0.028 

-5.8 % 

0.89 % 

OR_right 

0.086 ± 0.026  0.083 ± 0.022  0.089 ± 0.022  0.088 ± 0.030 

-3.77 % 

-0.46 % 

POPT_left 

0.121 ± 0.017  0.119 ± 0.019  0.117 ± 0.020  0.113 ± 0.033 

-1.27 % 

-3.2 % 

POPT_right 

0.113 ± 0.018  0.113 ± 0.018  0.111 ± 0.020  0.109 ± 0.029 

-0.26 % 

-1.29 % 

-0.140 

SCP_left 

0.114 ± 0.016  0.109 ± 0.015  0.100 ± 0.009  0.100 ± 0.019 

-4.04 % 

SCP_right 

0.104 ± 0.013  0.099 ± 0.014  0.097 ± 0.016  0.097 ± 0.017 

-4.74 % 

0.06 % 

-0.8 % 

SLF_III_left 

0.092 ± 0.015  0.086 ± 0.020  0.078 ± 0.012  0.070 ± 0.021 

-6.36 % 

-10.49 % 

SLF_III_right  0.083 ± 0.015  0.080 ± 0.015  0.080 ± 0.014  0.072 ± 0.015 

-3.91 % 

-9.31 % 

SLF_II_left 

0.079 ± 0.015  0.077 ± 0.020  0.068 ± 0.011  0.066 ± 0.022 

-2.78 % 

-4.03 % 

SLF_II_right 

0.072 ± 0.014  0.070 ± 0.015  0.065 ± 0.012  0.062 ± 0.017 

-2.79 % 

-5.41 % 

1.871 

2.750 

2.704 

1.660 

1.017 

1.003 

SLF_I_left 

0.083 ± 0.022  0.085 ± 0.025  0.087 ± 0.035  0.083 ± 0.047 

1.82 % 

SLF_I_right 

0.083 ± 0.020  0.084 ± 0.021  0.086 ± 0.027  0.082 ± 0.039 

0.49 % 

-5.55 % 

-4.34 % 

-0.631 

-0.327 

 313 

2.883 

2.333 

1.559 

2.462 

2.015 

2.161 

1.056 

0.069 

0.047 

0.356 

0.579 

0.755 

2.052 

1.210 

1.307 

0.874 

1.340 

1.345 

0.767 

1.257 

1.201 

2.999 

2.648 

3.386 

2.443 

0.665 

2.090 

0.754 

1.946 

0.967 

0.641 

0.027* 

0.033* 

0.098 

0.027* 

0.057 

0.047* 

0.16 

0.324 

0.306 

0.177 

0.325 

0.279 

0.226 

0.197 

0.055 

0.13 

0.12 

0.177 

0.116 

0.116 

0.197 

0.124 

0.13 

0.027* 

0.027* 

0.027* 

0.027* 

0.217 

0.053 

0.197 

0.06 

0.161 

0.219 

0.309 

0.068 

0.027* 

0.027* 

0.089 

0.161 

0.161 

0.219 

0.28 

1.056 

1.692 

2.417 

1.662 

0.060 

3.502 

2.090 

0.935 

0.221 

0.299 

0.196 

0.039 

0.963 

0.667 

0.151 

0.550 

0.129 

1.858 

1.539 

0.541 

1.298 

-0.233 

0.698 

1.628 

1.350 

0.548 

0.640 

-1.308 

0.516 

0.619 

0.073 

1.088 

0.607 

0.298 

0.296 

0.798 

1.678 

2.353 

0.510 

1.503 

0.454 

0.557 

0.501 

0.321 

0.313 

0.321 

0.606 

0.157 

0.313 

0.501 

0.577 

0.562 

0.581 

0.606 

0.501 

0.501 

0.59 

0.501 

0.59 

0.313 

0.343 

0.501 

0.417 

0.577 

0.501 

0.321 

0.417 

0.501 

0.501 

0.417 

0.501 

0.501 

0.606 

0.501 

0.501 

0.562 

0.562 

0.501 

0.321 

0.313 

0.501 

0.346 

0.514 

0.501 

 
 
 
 
 
 
 
 
Table A.25 (cont’d)  

TractSeg Name 

P1 

P2 

P1 

P2 

mTBI 

Controls 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

0.141 ± 
0.016 

0.131 ± 
0.017 

0.061 ± 
0.016 

0.066 ± 
0.017 

0.080 ± 
0.018 

0.073 ± 
0.020 

0.091 ± 
0.017 

0.087 ± 
0.018 

0.124 ± 
0.019 

0.111 ± 
0.019 

0.133 ± 
0.019 

0.122 ± 
0.018 

0.094 ± 
0.013 

0.095 ± 
0.013 

0.099 ± 
0.018 

0.104 ± 
0.016 

0.092 ± 
0.024 

0.085 ± 
0.026 

0.098 ± 
0.020 

0.097 ± 
0.021 

0.120 ± 
0.019 

0.105 ± 
0.019 

0.139 ± 
0.019 

0.120 ± 
0.018 

0.107 ± 
0.014 

0.106 ± 
0.016 

0.110 ± 
0.019 

0.136 ± 
0.026 

0.123 ± 
0.020 

0.057 ± 
0.023 

0.060 ± 
0.019 

0.074 ± 
0.017 

0.069 ± 
0.018 

0.089 ± 
0.018 

0.086 ± 
0.016 

0.121 ± 
0.025 

0.111 ± 
0.021 

0.128 ± 
0.026 

0.117 ± 
0.018 

0.091 ± 
0.020 

0.090 ± 
0.016 

0.094 ± 
0.027 

0.099 ± 
0.019 

0.086 ± 
0.022 

0.082 ± 
0.022 

0.096 ± 
0.021 

0.095 ± 
0.019 

0.119 ± 
0.025 

0.106 ± 
0.022 

0.135 ± 
0.028 

0.115 ± 
0.019 

0.103 ± 
0.022 

0.101 ± 
0.018 

0.106 ± 
0.029 

0.117 ± 
0.021 

0.119 ± 
0.025 

0.056 ± 
0.017 

0.066 ± 
0.021 

0.071 ± 
0.012 

0.070 ± 
0.014 

0.090 ± 
0.020 

0.086 ± 
0.016 

0.117 ± 
0.030 

0.115 ± 
0.029 

0.113 ± 
0.019 

0.115 ± 
0.022 

0.083 ± 
0.015 

0.088 ± 
0.018 

0.083 ± 
0.017 

0.096 ± 
0.019 

0.088 ± 
0.016 

0.088 ± 
0.022 

0.100 ± 
0.024 

0.098 ± 
0.025 

0.120 ± 
0.038 

0.114 ± 
0.039 

0.120 ± 
0.023 

0.114 ± 
0.025 

0.096 ± 
0.019 

0.100 ± 
0.023 

0.098 ± 
0.022 

0.113 ± 
0.030 

0.111 ± 
0.027 

0.053 ± 
0.022 

0.056 ± 
0.018 

0.072 ± 
0.020 

0.071 ± 
0.023 

0.086 ± 
0.030 

0.083 ± 
0.026 

0.109 ± 
0.038 

0.110 ± 
0.040 

0.109 ± 
0.031 

0.108 ± 
0.027 

0.074 ± 
0.016 

0.079 ± 
0.017 

0.071 ± 
0.017 

0.084 ± 
0.023 

0.088 ± 
0.028 

0.087 ± 
0.030 

0.095 ± 
0.037 

0.095 ± 
0.034 

0.111 ± 
0.047 

0.110 ± 
0.049 

0.116 ± 
0.037 

0.108 ± 
0.031 

0.086 ± 
0.021 

0.091 ± 
0.024 

0.086 ± 
0.027 

 314 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-3.47 % 

-4.08 % 

1.562 

0.098 

0.509 

0.501 

-5.49 % 

-7.14 % 

2.712 

0.027* 

0.898 

0.501 

-5.49 % 

-5.37 % 

1.381 

0.114 

0.671 

0.501 

-7.66 % 

-14.49 % 

2.478 

0.027* 

2.055 

0.313 

-6.34 % 

1.42 % 

2.544 

0.027* 

0.026 

0.606 

-5.05 % 

1.85 % 

1.527 

0.098 

0.964 

0.501 

-2.12 % 

-4.22 % 

0.971 

0.161 

0.644 

0.501 

-1.01 % 

-3.16 % 

0.176 

0.306 

0.600 

0.501 

-2.35 % 

-6.97 % 

0.972 

0.161 

0.802 

0.501 

-0.23 % 

-3.84 % 

0.191 

0.306 

0.467 

0.514 

-3.7 % 

-3.51 % 

1.619 

0.093 

0.558 

0.501 

-3.55 % 

-5.86 % 

1.702 

0.085 

1.028 

0.501 

-4.1 % 

-10.22 % 

1.740 

0.082 

1.974 

0.313 

-4.96 % 

-10.48 % 

2.735 

0.027* 

1.950 

0.313 

-4.37 % 

-14.08 % 

1.393 

0.114 

2.631 

0.313 

-4.73 % 

-12.23 % 

2.483 

0.027* 

2.250 

0.313 

-5.71 % 

0.41 % 

1.952 

0.06 

0.142 

0.59 

-3.74 % 

-0.36 % 

0.991 

0.161 

1.139 

0.501 

-1.45 % 

-4.59 % 

0.582 

0.226 

0.629 

0.501 

-1.29 % 

-3.7 % 

0.325 

0.28 

0.583 

0.501 

-1.31 % 

-7.8 % 

0.550 

0.229 

0.757 

0.501 

0.56 % 

-3.25 % 

-0.169 

0.306 

0.282 

0.562 

-3.17 % 

-3.68 % 

1.403 

0.114 

0.513 

0.501 

-4.04 % 

-5.42 % 

1.755 

0.082 

0.767 

0.501 

-3.38 % 

-10.47 % 

1.488 

0.102 

1.866 

0.313 

-5.11 % 

-9.32 % 

2.600 

0.027* 

1.682 

0.321 

-3.74 % 

-12.54 % 

1.289 

0.12 

1.917 

0.313 

 
 
 
 
 
 
 
Table A.25 (cont’d)  

TractSeg Name 

P1 

P2 

P1 

P2 

mTBI 

Controls 

T_PREM_right 

UF_left 

0.111 ± 
0.018 

0.043 ± 
0.015 

0.105 ± 
0.021 

0.040 ± 
0.016 

0.106 ± 
0.029 

0.041 ± 
0.012 

0.094 ± 
0.031 

0.039 ± 
0.019 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-5.76 % 

-10.69 % 

2.538 

0.027* 

1.611 

0.321 

-6.96 % 

-5.22 % 

1.531 

0.098 

0.317 

0.562 

UF_right 

0.051 ± 
0.014 

0.046 ± 
0.014 

0.046 ± 
0.022 
Tables A.25- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

0.050 ± 
0.019 

-8.54 % 

-8.95 % 

0.027* 

2.863 

0.692 

0.501 

 315 

 
 
 
 
 
 
 
 
 
 
 
0.828 

0.828 

0.869 

0.828 

0.828 

0.828 

0.828 

0.828 

0.832 

0.832 

0.832 

0.828 

0.832 

0.828 

0.828 

0.828 

0.828 

0.828 

0.828 

0.828 

0.828 

0.853 

0.832 

0.828 

0.869 

0.828 

0.869 

0.843 

0.828 

0.828 

0.828 

0.107 

-0.275 

-0.514 

-1.580 

-0.854 

-0.011 

-0.544 

-0.417 

-0.552 

-0.579 

0.683 

0.214 

-0.215 

-0.510 

-1.021 

-1.223 

-1.239 

-1.207 

-1.592 

-0.671 

-0.717 

-0.950 

-0.218 

-0.049 

-0.749 

-0.404 

-1.026 

-0.715 

0.578 

0.807 

-1.076 

0.801 

0.793 

0.776 

0.776 

0.776 

0.824 

0.776 

0.776 

0.776 

0.776 

0.776 

0.797 

0.797 

0.776 

0.776 

0.776 

0.776 

0.776 

0.776 

0.776 

0.776 

0.776 

0.797 

0.817 

0.776 

0.776 

0.776 

0.776 

0.776 

0.776 

0.776 

Table A.26 Results of Post Hoc Tract Specific Comparison of Fixel Based Analysis Fiber Density 

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

0.285 ± 0.018  0.286 ± 0.014 

0.296 ± 0.014 

0.296 ± 0.011 

0.39 % 

-0.12 % 

-0.733 

ATR_right 

0.291 ± 0.014  0.292 ± 0.015 

0.306 ± 0.014 

0.310 ± 0.014 

0.294 ± 0.015  0.293 ± 0.014 

0.297 ± 0.015 

0.297 ± 0.013 

-0.23 % 

0.298 ± 0.014  0.298 ± 0.016 

0.316 ± 0.016 

0.318 ± 0.014 

0 % 

0.18 % 

0.81 % 

0.269 ± 0.025  0.271 ± 0.024 

0.295 ± 0.027 

0.302 ± 0.025 

0.378 ± 0.024  0.374 ± 0.026 

0.392 ± 0.024 

0.393 ± 0.023 

-1.04 % 

0.331 ± 0.018  0.330 ± 0.019 

0.344 ± 0.018 

0.346 ± 0.015 

-0.21 % 

0.326 ± 0.022  0.328 ± 0.024 

0.344 ± 0.029 

0.346 ± 0.028 

0.6 % 

0.366 ± 0.023  0.367 ± 0.023 

0.386 ± 0.020 

0.389 ± 0.019 

0.14 % 

0.370 ± 0.021  0.370 ± 0.021 

0.385 ± 0.025 

0.389 ± 0.024 

-0.11 % 

0.28 % 

0.57 % 

1.4 % 

2.26 % 

0.02 % 

0.55 % 

0.55 % 

0.6 % 

0.8 % 

0.783 

0.011 

-0.516 

-1.089 

1.933 

0.624 

-1.192 

-0.348 

0.294 

0.318 

AF_left 

AF_right 

ATR_left 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

0.352 ± 0.023  0.352 ± 0.019 

0.359 ± 0.015 

0.357 ± 0.016 

-0.15 % 

-0.55 % 

0.330 ± 0.029  0.331 ± 0.020 

0.333 ± 0.014 

0.332 ± 0.016 

0.343 ± 0.018  0.343 ± 0.016 

0.353 ± 0.016 

0.354 ± 0.016 

0.4 % 

-0.1 % 

0.314 ± 0.016  0.313 ± 0.017 

0.324 ± 0.016 

0.326 ± 0.013 

-0.24 % 

0.314 ± 0.017  0.313 ± 0.018 

0.319 ± 0.019 

0.322 ± 0.014 

-0.33 % 

0.428 ± 0.016  0.431 ± 0.016 

0.449 ± 0.017 

0.454 ± 0.017 

0.437 ± 0.016  0.437 ± 0.014 

0.451 ± 0.015 

0.455 ± 0.018 

0.398 ± 0.015  0.400 ± 0.015 

0.416 ± 0.015 

0.420 ± 0.016 

0.400 ± 0.014  0.401 ± 0.013 

0.413 ± 0.016 

0.418 ± 0.016 

0.210 ± 0.052  0.213 ± 0.057 

0.237 ± 0.051 

0.241 ± 0.056 

0.221 ± 0.054  0.224 ± 0.059 

0.236 ± 0.058 

0.241 ± 0.055 

0.302 ± 0.017  0.302 ± 0.012 

0.302 ± 0.014 

0.305 ± 0.015 

0.277 ± 0.016  0.277 ± 0.015 

0.277 ± 0.010 

0.278 ± 0.017 

0.313 ± 0.017  0.314 ± 0.015 

0.323 ± 0.012 

0.323 ± 0.013 

0.313 ± 0.016  0.313 ± 0.014 

0.319 ± 0.011 

0.321 ± 0.012 

0.288 ± 0.022  0.290 ± 0.015 

0.293 ± 0.013 

0.295 ± 0.013 

0.291 ± 0.019  0.291 ± 0.016 

0.296 ± 0.012 

0.299 ± 0.013 

0.315 ± 0.014  0.315 ± 0.012 

0.315 ± 0.011 

0.317 ± 0.010 

0.308 ± 0.022  0.308 ± 0.016 

0.315 ± 0.016 

0.313 ± 0.018 

0.67 % 

0.18 % 

0.51 % 

0.24 % 

1.45 % 

1.2 % 

0.1 % 

-0.2 % 

0.15 % 

0.03 % 

0.84 % 

0.02 % 

0.1 % 

0.3 % 

-0.17 % 

-0.513 

0.2 % 

0.58 % 

0.95 % 

1.04 % 

0.95 % 

1.06 % 

1.16 % 

1.87 % 

2.37 % 

0.95 % 

0.25 % 

0.05 % 

0.57 % 

0.37 % 

1.01 % 

0.52 % 

0.288 

0.585 

0.994 

-1.502 

-0.540 

-1.399 

-0.812 

-1.221 

-0.978 

-0.168 

0.281 

-0.398 

-0.088 

-1.323 

-0.057 

-0.215 

-0.57 % 

-0.566 

0.325 ± 0.018  0.324 ± 0.017 

0.328 ± 0.015 

0.325 ± 0.017 

-0.46 % 

-0.75 % 

1.367 

0.316 ± 0.021  0.318 ± 0.021 

0.325 ± 0.014 

0.328 ± 0.014 

0.49 % 

0.92 % 

-1.147 

 316 

 
 
 
 
 
 
 
 
Table A.26 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_rig
ht 

ST_PREC_left 

ST_PREC_right 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_righ
t 

0.335 ± 
0.022 

0.379 ± 
0.015 

0.402 ± 
0.017 

0.342 ± 
0.012 

0.339 ± 
0.011 

0.304 ± 
0.017 

0.300 ± 
0.018 

0.284 ± 
0.017 

0.281 ± 
0.017 

0.276 ± 
0.014 

0.289 ± 
0.015 

0.416 ± 
0.017 

0.434 ± 
0.015 

0.320 ± 
0.019 

0.335 ± 
0.021 

0.326 ± 
0.023 

0.334 ± 
0.022 

0.335 ± 
0.016 

0.355 ± 
0.017 

0.338 ± 
0.013 

0.361 ± 
0.015 

0.346 ± 
0.013 

0.363 ± 
0.014 

0.329 ± 
0.013 

0.330 ± 
0.013 

0.308 ± 
0.014 

0.322 ± 
0.016 

0.336 ± 
0.019 

0.380 ± 
0.014 

0.402 ± 
0.016 

0.344 ± 
0.012 

0.340 ± 
0.012 

0.305 ± 
0.015 

0.299 ± 
0.018 

0.284 ± 
0.017 

0.281 ± 
0.018 

0.275 ± 
0.014 

0.289 ± 
0.015 

0.418 ± 
0.018 

0.433 ± 
0.015 

0.319 ± 
0.019 

0.334 ± 
0.019 

0.327 ± 
0.020 

0.335 ± 
0.018 

0.335 ± 
0.015 

0.354 ± 
0.016 

0.340 ± 
0.014 

0.360 ± 
0.014 

0.348 ± 
0.014 

0.363 ± 
0.013 

0.329 ± 
0.015 

0.329 ± 
0.012 

0.309 ± 
0.016 

0.324 ± 
0.016 

0.338 ± 
0.012 

0.390 ± 
0.018 

0.411 ± 
0.015 

0.351 ± 
0.013 

0.344 ± 
0.011 

0.313 ± 
0.019 

0.298 ± 
0.017 

0.288 ± 
0.021 

0.283 ± 
0.013 

0.278 ± 
0.016 

0.295 ± 
0.014 

0.432 ± 
0.022 

0.444 ± 
0.023 

0.339 ± 
0.015 

0.350 ± 
0.016 

0.335 ± 
0.013 

0.339 ± 
0.013 

0.345 ± 
0.016 

0.361 ± 
0.013 

0.355 ± 
0.017 

0.374 ± 
0.016 

0.365 ± 
0.015 

0.376 ± 
0.015 

0.344 ± 
0.013 

0.340 ± 
0.013 

0.320 ± 
0.024 

0.335 ± 
0.019 

0.340 ± 
0.014 

0.391 ± 
0.019 

0.412 ± 
0.018 

0.356 ± 
0.012 

0.348 ± 
0.013 

0.314 ± 
0.019 

0.299 ± 
0.017 

0.289 ± 
0.020 

0.283 ± 
0.015 

0.278 ± 
0.017 

0.294 ± 
0.015 

0.439 ± 
0.020 

0.449 ± 
0.023 

0.341 ± 
0.018 

0.355 ± 
0.016 

0.337 ± 
0.015 

0.341 ± 
0.015 

0.345 ± 
0.018 

0.361 ± 
0.016 

0.358 ± 
0.018 

0.376 ± 
0.019 

0.368 ± 
0.016 

0.379 ± 
0.015 

0.345 ± 
0.015 

0.344 ± 
0.013 

0.322 ± 
0.023 

0.337 ± 
0.017 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

0.35 % 

0.74 % 

-0.752 

0.828 

-1.054 

0.776 

0.2 % 

0.4 % 

-0.546 

0.828 

-0.478 

0.776 

0 % 

0.29 % 

-0.008 

0.869 

-0.417 

0.776 

0.44 % 

1.66 % 

-1.511 

0.828 

-2.619 

0.776 

0.15 % 

1.2 % 

-0.425 

0.828 

-1.691 

0.776 

0.1 % 

0.2 % 

-0.228 

0.843 

-0.173 

0.797 

-0.29 % 

0.16 % 

1.069 

0.828 

-0.156 

0.797 

0.1 % 

0.37 % 

-0.259 

0.836 

-0.337 

0.779 

-0.19 % 

-0.12 % 

0.568 

0.828 

0.146 

0.797 

-0.26 % 

-0.15 % 

0.811 

0.828 

0.113 

0.801 

0 % 

-0.39 % 

0.010 

0.869 

0.375 

0.779 

0.64 % 

1.65 % 

-1.996 

0.828 

-2.003 

0.776 

-0.02 % 

1.16 % 

0.072 

0.869 

-1.496 

0.776 

-0.29 % 

0.64 % 

0.543 

0.828 

-0.417 

0.776 

-0.52 % 

1.24 % 

1.006 

0.828 

-1.223 

0.776 

0.46 % 

0.59 % 

-0.986 

0.828 

-0.617 

0.776 

0.32 % 

0.45 % 

-0.669 

0.828 

-0.622 

0.776 

0.18 % 

0.08 % 

-0.497 

0.828 

-0.084 

0.805 

-0.23 % 

0 % 

0.741 

0.828 

-0.002 

0.824 

0.54 % 

0.85 % 

-1.536 

0.828 

-0.728 

0.776 

-0.2 % 

0.53 % 

0.667 

0.828 

-0.492 

0.776 

0.57 % 

0.87 % 

-1.420 

0.828 

-0.802 

0.776 

0.13 % 

0.75 % 

-0.406 

0.828 

-0.792 

0.776 

0.04 % 

0.51 % 

-0.120 

0.869 

-0.452 

0.776 

-0.12 % 

1.13 % 

0.442 

0.828 

-1.413 

0.776 

0.38 % 

0.64 % 

-0.876 

0.828 

-0.455 

0.776 

0.53 % 

0.61 % 

-1.384 

0.828 

-0.507 

0.776 

 317 

 
 
 
 
 
 
 
Table A.26 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

T_POSTC_left 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

UF_right 

0.310 ± 
0.021 

0.328 ± 
0.021 

0.335 ± 
0.015 

0.362 ± 
0.017 

0.357 ± 
0.014 

0.372 ± 
0.017 

0.365 ± 
0.015 

0.377 ± 
0.015 

0.333 ± 
0.013 

0.322 ± 
0.012 

0.324 ± 
0.013 

0.335 ± 
0.015 

0.299 ± 
0.020 

0.295 ± 
0.016 

P2 

0.312 ± 
0.020 

0.329 ± 
0.019 

0.335 ± 
0.015 

0.361 ± 
0.017 

0.358 ± 
0.015 

0.371 ± 
0.017 

0.367 ± 
0.015 

0.377 ± 
0.014 

0.333 ± 
0.014 

0.322 ± 
0.013 

0.326 ± 
0.016 

0.337 ± 
0.017 

0.299 ± 
0.017 

0.293 ± 
0.017 

P1 

P2 

0.319 ± 
0.013 

0.331 ± 
0.012 

0.344 ± 
0.017 

0.367 ± 
0.015 

0.371 ± 
0.021 

0.383 ± 
0.020 

0.386 ± 
0.016 

0.390 ± 
0.016 

0.349 ± 
0.013 

0.334 ± 
0.013 

0.340 ± 
0.021 

0.352 ± 
0.019 

0.313 ± 
0.015 

0.307 ± 
0.011 

0.322 ± 
0.014 

0.333 ± 
0.014 

0.345 ± 
0.018 

0.368 ± 
0.017 

0.374 ± 
0.021 

0.387 ± 
0.021 

0.390 ± 
0.016 

0.393 ± 
0.016 

0.352 ± 
0.014 

0.338 ± 
0.013 

0.343 ± 
0.020 

0.356 ± 
0.018 

0.310 ± 
0.024 

0.307 ± 
0.020 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

0.46 % 

0.96 % 

t-score 
mTBI 

-1.053 

q-value 
mTBI1 

0.828 

t-score 
controls 

-1.122 

q-value 
controls 

0.776 

0.37 % 

0.61 % 

-0.774 

0.828 

-0.859 

0.776 

0.06 % 

0.29 % 

-0.169 

0.853 

-0.329 

0.779 

-0.17 % 

0.24 % 

0.545 

0.828 

-0.298 

0.789 

0.25 % 

1.04 % 

-0.735 

0.828 

-1.154 

0.776 

-0.27 % 

0.86 % 

0.846 

0.828 

-0.903 

0.776 

0.52 % 

1 % 

-1.303 

0.828 

-1.008 

0.776 

0.13 % 

0.94 % 

-0.392 

0.828 

-1.137 

0.776 

0.14 % 

0.69 % 

-0.430 

0.828 

-0.673 

0.776 

0 % 

1.19 % 

0.018 

0.869 

-1.470 

0.776 

0.5 % 

0.78 % 

-1.164 

0.828 

-0.619 

0.776 

0.63 % 

0.94 % 

-1.532 

0.828 

-0.896 

0.776 

-0.16 % 

-0.77 % 

0.300 

0.832 

0.344 

0.779 

-0.41 % 

0.27 % 

0.868 

0.828 

-0.152 

0.797 

Tables A.26- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

 318 

 
 
 
 
 
 
 
 
 
 
 
 
Table A.27 Results of Post Hoc Tract Specific Comparison of Fixel Based Analysis Fiber Cross 
Section 

mTBI 

Controls 

TractSeg 
Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI 
% 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

AF_left 

0.035 ± 0.073 

0.035 ± 0.070 

0.056 ± 0.111 

0.068 ± 0.125 

-0.26 %  21.44 % 

0.021 

0.16 

AF_right 

0.046 ± 0.070 

0.045 ± 0.073 

0.078 ± 0.101 

0.087 ± 0.121 

-1.42 %  11.55 % 

0.231 

ATR_left 

0.003 ± 0.082 

-6.888e-03 ± 
8.403e-02 

0.026 ± 0.102 

0.028 ± 0.121 

ATR_right 

0.014 ± 0.079 

0.008 ± 0.080 

0.052 ± 0.091 

0.056 ± 0.106 

CA 

0.005 ± 0.083 

0.010 ± 0.072 

0.056 ± 0.089 

0.073 ± 0.085 

-299.02 
% 

-43.33 
% 

104.23 
% 

-1.356 

-1.252 

-0.194 

0.085 

0.087 

0.181 

0.147 

0.061 

8.83 % 

1.933 

6.57 % 

1.259 

0.078 

-0.610 

0.135 

29.87 % 

-0.943 

0.108 

-1.902 

0.06 

286.67 
% 

-75.08 
% 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

0.013 ± 0.089 

0.003 ± 0.092 

0.077 ± 0.107 

0.081 ± 0.101 

-75.6 % 

4.94 % 

0.020 ± 0.069 

0.015 ± 0.071 

0.054 ± 0.093 

0.057 ± 0.102 

-28.35 
% 

6.43 % 

1.615 

1.994 

0.065 

0.061 

-0.522 

-0.942 

0.023 ± 0.080 

0.026 ± 0.088 

0.028 ± 0.103 

0.039 ± 0.119 

10.96 % 

39.9 % 

-0.437 

0.142 

-1.329 

0.003 ± 0.072 

0.013 ± 0.075 

0.047 ± 0.078 

0.064 ± 0.097 

37.54 % 

-1.944 

0.061 

-1.859 

0.145 

0.109 

0.086 

0.06 

-1.106e-02 ± 
6.943e-02 

8.253e-04 ± 
7.889e-02 

-2.758e-03 ± 
7.171e-02 

0.007 ± 0.103 

0.017 ± 0.113 

156.28 % 

-1.901 

0.061 

-1.746 

0.066 

0.006 ± 0.073 

0.006 ± 0.106 

0.014 ± 0.110 

612 % 

122.82 % 

-0.844 

0.113 

-2.446 

0.06 

0.057 ± 0.079 

0.053 ± 0.080 

0.043 ± 0.103 

0.060 ± 0.115 

-6.36 %  40.95 % 

1.267 

0.078 

-2.471 

0.013 ± 0.061 

0.014 ± 0.063 

0.032 ± 0.091 

0.040 ± 0.099 

8.12 % 

25.81 % 

-0.333 

0.147 

-2.404 

0.06 

0.06 

CG_left 

0.008 ± 0.063 

0.006 ± 0.063 

0.026 ± 0.101 

0.022 ± 0.092 

CG_right 

0.008 ± 0.063 

0.004 ± 0.064 

0.024 ± 0.094 

0.025 ± 0.090 

CST_left 

CST_right 

-1.905e-02 ± 
7.521e-02 

-3.508e-03 ± 
7.713e-02 

-8.765e-03 ± 
7.238e-02 

0.039 ± 0.085 

0.051 ± 0.095 

0.003 ± 0.076 

0.035 ± 0.093 

0.041 ± 0.096 

-26.31 
% 

-47.98 
% 

-53.99 
% 

-174.57 
% 

-14.76 % 

0.728 

0.128 

0.492 

0.145 

5.64 % 

1.359 

0.078 

-0.181 

0.181 

29.98 % 

-2.866  0.022* 

-1.627 

0.07 

17.98 % 

-1.265 

0.078 

-0.834 

0.114 

0.08 

0.099 

FPT_left 

0.004 ± 0.068 

0.006 ± 0.065 

0.032 ± 0.073 

0.039 ± 0.081 

52.09 %  23.18 % 

-0.679 

0.133 

-1.465 

FPT_right 

0.001 ± 0.064 

FX_left 

FX_right 

-7.329e-02 ± 
1.008e-01 

-7.471e-02 ± 
1.073e-01 

4.733e-04 ± 
6.411e-02 

-7.435e-02 ± 
1.209e-01 

-7.727e-02 ± 
1.254e-01 

0.030 ± 0.091 

0.033 ± 0.090 

-65.91 
% 

11.89 % 

0.382 

0.142 

-1.055 

-1.159e-01 ± 
1.119e-01 

-1.160e-01 ± 
1.090e-01 

-1.177e-01 ± 
1.174e-01 

-1.159e-01 ± 
1.192e-01 

1.44 % 

1.52 % 

0.210 

0.148 

0.366 

0.163 

3.42 % 

-0.08 % 

0.525 

0.142 

-0.014 

0.202 

ICP_left 

0.025 ± 0.091 

0.024 ± 0.097 

0.054 ± 0.097 

0.049 ± 0.090 

-4.04 % 

-9.26 % 

0.192 

0.148 

0.489 

ICP_right 

0.023 ± 0.081 

0.018 ± 0.085 

0.036 ± 0.094 

0.038 ± 0.103 

-24.1 % 

7.15 % 

IFO_left 

0.035 ± 0.074 

0.028 ± 0.074 

0.044 ± 0.105 

0.055 ± 0.120 

-18.8 % 

23 % 

1.047 

1.730 

IFO_right 

0.047 ± 0.067 

0.042 ± 0.068 

0.066 ± 0.099 

0.077 ± 0.111 

-10.62 
% 

16.59 % 

1.649 

ILF_left 

0.069 ± 0.085 

0.067 ± 0.087 

0.070 ± 0.107 

0.091 ± 0.113 

-2.51 %  28.83 % 

0.397 

ILF_right 

0.091 ± 0.082 

0.090 ± 0.082 

0.091 ± 0.104 

0.107 ± 0.107 

-1 % 

18.01 % 

0.263 

MCP 

0.028 ± 0.094 

0.027 ± 0.097 

0.044 ± 0.105 

0.045 ± 0.107 

-5.45 % 

0.86 % 

0.289 

0.098 

0.065 

0.065 

0.142 

0.147 

0.147 

-0.207 

-1.608 

-1.843 

-3.214 

-1.738 

-0.037 

0.145 

0.181 

0.07 

0.06 

0.06 

0.066 

0.202 

 319 

 
 
 
 
 
 
 
 
Table A.27 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

MLF_left 

MLF_right 

OR_left 

OR_right 

POPT_left 

POPT_right 

SCP_left 

SCP_right 

SLF_III_left 

SLF_III_right 

SLF_II_left 

SLF_II_right 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

ST_FO_left 

ST_FO_right 

ST_OCC_left 

ST_OCC_right 

ST_PAR_left 

ST_PAR_right 

ST_POSTC_left 

ST_POSTC_right 

ST_PREC_left 

ST_PREC_right 

0.017 ± 
0.082 

0.020 ± 
0.078 

0.037 ± 
0.080 

0.035 ± 
0.075 

0.020 ± 
0.075 

0.025 ± 
0.073 

0.038 ± 
0.080 

0.034 ± 
0.074 

0.026 ± 
0.110 

0.026 ± 
0.102 

0.036 ± 
0.098 

0.042 ± 
0.104 

0.034 ± 
0.113 

0.032 ± 
0.108 

0.052 ± 
0.114 

0.055 ± 
0.118 

-3.043e-02 ± 
7.537e-02 

-2.240e-02 ± 
6.974e-02 

-6.398e-03 ± 
9.750e-02 

-4.521e-04 ± 
1.052e-01 

-2.630e-02 ± 
7.626e-02 

-1.965e-02 ± 
7.247e-02 

-4.322e-03 ± 
1.042e-01 

-9.075e-05 ± 
1.045e-01 

0.007 ± 
0.080 

0.017 ± 
0.072 

0.036 ± 
0.081 

0.047 ± 
0.074 

0.027 ± 
0.072 

0.035 ± 
0.079 

0.007 ± 
0.069 

0.006 ± 
0.074 

-7.186e-03 ± 
7.695e-02 

0.006 ± 
0.065 

0.010 ± 
0.081 

0.033 ± 
0.080 

0.042 ± 
0.078 

0.045 ± 
0.074 

0.008 ± 
0.085 

0.014 ± 
0.073 

0.036 ± 
0.082 

0.049 ± 
0.078 

0.028 ± 
0.073 

0.033 ± 
0.078 

0.015 ± 
0.067 

0.008 ± 
0.072 

0.002 ± 
0.075 

0.010 ± 
0.063 

5.806e-06 ± 
8.671e-02 

0.026 ± 
0.081 

0.042 ± 
0.079 

0.044 ± 
0.074 

-1.348e-02 ± 
7.252e-02 

-6.068e-03 ± 
7.133e-02 

-1.055e-02 ± 
7.394e-02 

-4.638e-03 ± 
7.093e-02 

-2.821e-02 ± 
6.992e-02 

-1.716e-02 ± 
7.083e-02 

-1.800e-02 ± 
7.040e-02 

-1.046e-02 ± 
7.192e-02 

-6.573e-03 ± 
6.506e-02 

0.002 ± 
0.068 

0.004 ± 
0.066 

0.008 ± 
0.069 

0.032 ± 
0.090 

0.033 ± 
0.087 

0.062 ± 
0.128 

0.089 ± 
0.115 

0.043 ± 
0.105 

0.060 ± 
0.104 

0.025 ± 
0.091 

0.026 ± 
0.110 

0.029 ± 
0.096 

0.030 ± 
0.108 

0.038 ± 
0.106 

0.079 ± 
0.092 

0.044 ± 
0.097 

0.050 ± 
0.103 

0.009 ± 
0.103 

0.010 ± 
0.106 

0.012 ± 
0.102 

0.016 ± 
0.113 

0.049 ± 
0.085 

0.035 ± 
0.096 

0.029 ± 
0.086 

0.035 ± 
0.093 

0.067 ± 
0.134 

0.093 ± 
0.124 

0.058 ± 
0.134 

0.071 ± 
0.120 

0.036 ± 
0.099 

0.035 ± 
0.110 

0.041 ± 
0.110 

0.035 ± 
0.106 

0.043 ± 
0.114 

0.094 ± 
0.094 

0.062 ± 
0.115 

0.066 ± 
0.117 

0.018 ± 
0.115 

0.018 ± 
0.114 

0.024 ± 
0.123 

0.023 ± 
0.124 

0.064 ± 
0.103 

0.044 ± 
0.108 

 320 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

18.51 % 

32.39 % 

-0.505 

0.142 

-2.264 

0.06 

28.69 % 

24.39 % 

-1.063 

0.098 

-1.374 

0.085 

1.13 % 

44.33 % 

-0.122 

0.153 

-2.059 

0.06 

-1.84 % 

30.6 % 

0.255 

0.147 

-1.947 

0.06 

-26.39 % 

-92.93 % 

-1.630 

0.065 

-1.250 

0.087 

-25.31 % 

-97.9 % 

-1.323 

0.078 

-0.685 

0.126 

28.33 % 

-8.05 % 

-0.543 

0.142 

0.332 

0.166 

-15.59 % 

4.87 % 

0.837 

0.113 

-0.179 

0.181 

-1 % 

6.67 % 

0.093 

0.153 

-0.521 

0.145 

5.05 % 

4.68 % 

-0.870 

0.113 

-1.076 

0.098 

4.02 % 

37.08 % 

-0.393 

0.142 

-1.245 

0.087 

-6.27 % 

17.78 % 

0.643 

0.135 

-1.606 

0.07 

103.82 % 

45.42 % 

-1.627 

0.065 

-2.066 

0.06 

34.8 % 

36.4 % 

-0.512 

0.142 

-2.073 

0.06 

-127.13 
% 

40.41 % 

-2.528 

0.028* 

-1.510 

0.077 

63.84 % 

16.85 % 

-0.839 

0.113 

-0.734 

0.122 

-99.94 % 

14.79 % 

1.625 

0.065 

-0.733 

0.122 

-20.48 % 

18.12 % 

0.947 

0.108 

-2.621 

0.06 

-0.72 % 

42.71 % 

0.100 

0.153 

-2.449 

0.06 

-3.77 % 

30.42 % 

0.634 

0.135 

-2.194 

0.06 

-54.99 % 

95 % 

-1.546 

0.068 

-1.580 

0.071 

-56.06 % 

69.63 % 

-1.326 

0.078 

-1.135 

0.096 

-39.16 % 

96.42 % 

-2.625 

0.026* 

-1.294 

0.087 

-41.9 % 

40.32 % 

-2.291 

0.043* 

-0.744 

0.122 

-163.74 
% 

31.1 % 

-3.168 

0.014* 

-1.844 

0.06 

299.6 % 

25.28 % 

-1.343 

0.078 

-1.089 

0.098 

 
 
 
 
 
 
 
Table A.27 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

0.016 ± 
0.067 

0.020 ± 
0.065 

0.012 ± 
0.074 

0.016 ± 
0.070 

0.038 ± 
0.079 

0.037 ± 
0.075 

P2 

0.012 ± 
0.067 

0.015 ± 
0.065 

0.015 ± 
0.076 

0.018 ± 
0.074 

0.039 ± 
0.080 

0.036 ± 
0.075 

P1 

0.047 ± 
0.085 

0.060 ± 
0.093 

0.028 ± 
0.094 

0.041 ± 
0.094 

0.036 ± 
0.098 

0.043 ± 
0.105 

0.003 ± 
0.101 

ST_PREF_left 

ST_PREF_right 

ST_PREM_left 

ST_PREM_right 

T_OCC_left 

T_OCC_right 

T_PAR_left 

T_PAR_right 

-1.725e-02 ± 
7.338e-02 

-9.715e-03 ± 
7.206e-02 

-1.994e-02 ± 
7.469e-02 

-1.372e-02 ± 
7.168e-02 

-1.538e-03 ± 
1.027e-01 

T_POSTC_left 

-3.074e-02 ± 
7.699e-02 

-2.048e-02 ± 
7.665e-02 

8.301e-04 ± 
1.098e-01 

T_POSTC_right 

T_PREC_left 

T_PREC_right 

T_PREF_left 

T_PREF_right 

T_PREM_left 

T_PREM_right 

UF_left 

-1.552e-02 ± 
7.574e-02 

-7.507e-03 ± 
7.689e-02 

-9.899e-03 ± 
6.619e-02 

6.543e-04 ± 
6.824e-02 

0.003 ± 
0.070 

0.014 ± 
0.067 

0.016 ± 
0.066 

0.006 ± 
0.071 

0.012 ± 
0.070 

0.025 ± 
0.077 

0.009 ± 
0.072 

0.010 ± 
0.068 

0.011 ± 
0.067 

0.008 ± 
0.074 

0.014 ± 
0.074 

0.016 ± 
0.079 

0.014 ± 
0.119 

0.049 ± 
0.083 

0.037 ± 
0.094 

0.041 ± 
0.084 

0.050 ± 
0.093 

0.020 ± 
0.090 

0.033 ± 
0.089 

0.047 ± 
0.112 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls 
% 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

-24.73 % 

14.88 % 

1.240 

0.078 

-0.922 

0.109 

-24.18 % 

8.64 % 

1.859 

0.061 

-1.377 

0.085 

19.98 % 

41.08 % 

-0.589 

0.139 

-0.864 

0.114 

12.14 % 

24.11 % 

-0.447 

0.142 

-1.644 

0.07 

1 % 

45.46 % 

-0.115 

0.153 

-2.107 

0.06 

-1.82 % 

30.71 % 

0.258 

0.147 

-1.976 

0.06 

-43.68 %  281.16 % 

-1.556 

0.068 

-1.389 

0.085 

-31.2 % 

-414.66 % 

-1.255 

0.078 

-0.962 

0.108 

-33.37 %  1118.82 % 

-2.217 

0.045* 

-1.235 

0.087 

-51.62 % 

43.72 % 

-2.685 

0.026* 

-0.723 

0.122 

-106.61 
% 

31.61 % 

-3.228 

0.014* 

-1.879 

0.06 

196.31 % 

25.4 % 

-1.356 

0.078 

-1.104 

0.098 

-25.47 % 

16.15 % 

1.137 

0.09 

-0.835 

0.114 

-29.73 % 

9.28 % 

1.707 

0.065 

-1.174 

0.093 

23.69 % 

46.58 % 

-0.405 

0.142 

-0.845 

0.114 

13.12 % 

32.52 % 

-0.432 

0.142 

-2.023 

0.06 

-33.31 % 

21.04 % 

1.480 

0.073 

-1.037 

0.099 

P2 

0.054 ± 
0.100 

0.065 ± 
0.102 

0.040 ± 
0.118 

0.050 ± 
0.105 

0.052 ± 
0.115 

0.056 ± 
0.119 

0.010 ± 
0.111 

0.005 ± 
0.106 

0.010 ± 
0.126 

0.021 ± 
0.126 

0.064 ± 
0.099 

0.046 ± 
0.102 

0.048 ± 
0.097 

0.054 ± 
0.103 

0.030 ± 
0.108 

0.044 ± 
0.096 

0.057 ± 
0.116 

UF_right 

0.034 ± 
0.071 

0.033 ± 
0.066 

0.087 ± 
0.096 
Tables A.27- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

0.074 ± 
0.086 

17.82 % 

-3.78 % 

-2.246 

0.147 

0.242 

0.06 

 321 

 
 
 
 
 
 
 
 
 
 
 
Table A.28 Results of Post Hoc Tract Specific Comparison of Fixel Based Analysis Fiber Density 
Cross Section 

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

AF_left 

AF_right 

ATR_left 

0.302 ± 0.033 

0.302 ± 0.029 

0.322 ± 0.045 

0.327 ± 0.043 

0.27 % 

0.314 ± 0.031 

0.312 ± 0.031 

0.327 ± 0.044 

0.333 ± 0.046 

-0.41 % 

0.304 ± 0.032 

0.301 ± 0.033 

0.332 ± 0.044 

0.335 ± 0.046 

-0.96 % 

ATR_right 

0.300 ± 0.030 

0.299 ± 0.031 

0.328 ± 0.036 

0.335 ± 0.042 

-0.38 % 

CA 

CC_1 

CC_2 

CC_3 

CC_4 

CC_5 

CC_6 

CC_7 

CC 

CG_left 

CG_right 

CST_left 

CST_right 

FPT_left 

FPT_right 

FX_left 

FX_right 

ICP_left 

ICP_right 

IFO_left 

IFO_right 

ILF_left 

ILF_right 

MCP 

MLF_left 

MLF_right 

OR_left 

OR_right 

0.274 ± 0.040 

0.277 ± 0.037 

0.316 ± 0.047 

0.329 ± 0.040 

1.14 % 

0.388 ± 0.051 

0.380 ± 0.051 

0.435 ± 0.068 

0.437 ± 0.058 

-2.08 % 

0.342 ± 0.031 

0.340 ± 0.032 

0.370 ± 0.045 

0.375 ± 0.043 

-0.68 % 

0.337 ± 0.032 

0.340 ± 0.036 

0.363 ± 0.053 

0.370 ± 0.056 

0.84 % 

0.371 ± 0.033 

0.375 ± 0.034 

0.408 ± 0.038 

0.420 ± 0.040 

1.1 % 

0.368 ± 0.032 

0.371 ± 0.034 

0.392 ± 0.044 

0.400 ± 0.041 

0.81 % 

0.360 ± 0.042 

0.360 ± 0.036 

0.369 ± 0.046 

0.370 ± 0.042 

0.12 % 

0.353 ± 0.046 

0.353 ± 0.039 

0.353 ± 0.042 

0.359 ± 0.043 

-0.03 % 

0.352 ± 0.030 

0.352 ± 0.029 

0.371 ± 0.041 

0.376 ± 0.038 

-0.04 % 

0.322 ± 0.030 

0.320 ± 0.030 

0.341 ± 0.051 

0.340 ± 0.036 

-0.63 % 

-0.14 % 

0.322 ± 0.031 

0.319 ± 0.032 

0.333 ± 0.048 

0.336 ± 0.038 

-0.8 % 

0.424 ± 0.042 

0.431 ± 0.040 

0.472 ± 0.049 

0.483 ± 0.047 

1.66 % 

0.440 ± 0.041 

0.443 ± 0.039 

0.473 ± 0.052 

0.480 ± 0.050 

0.71 % 

0.404 ± 0.037 

0.406 ± 0.036 

0.435 ± 0.042 

0.443 ± 0.042 

0.64 % 

0.404 ± 0.034 

0.405 ± 0.033 

0.432 ± 0.048 

0.439 ± 0.045 

0.21 % 

0.190 ± 0.044 

0.192 ± 0.049 

0.207 ± 0.046 

0.211 ± 0.051 

0.98 % 

0.199 ± 0.045 

0.200 ± 0.049 

0.206 ± 0.050 

0.211 ± 0.050 

0.56 % 

0.315 ± 0.040 

0.315 ± 0.041 

0.321 ± 0.034 

0.323 ± 0.029 

0.04 % 

0.287 ± 0.031 

0.285 ± 0.032 

0.290 ± 0.031 

0.291 ± 0.030 

-0.65 % 

0.328 ± 0.032 

0.326 ± 0.032 

0.342 ± 0.042 

0.347 ± 0.042 

-0.44 % 

0.332 ± 0.030 

0.330 ± 0.029 

0.345 ± 0.039 

0.351 ± 0.042 

-0.5 % 

0.311 ± 0.038 

0.313 ± 0.035 

0.320 ± 0.044 

0.327 ± 0.040 

0.64 % 

0.322 ± 0.035 

0.322 ± 0.035 

0.329 ± 0.042 

0.337 ± 0.036 

-0.01 % 

0.330 ± 0.040 

0.330 ± 0.040 

0.333 ± 0.039 

0.336 ± 0.037 

-0.03 % 

0.319 ± 0.037 

0.320 ± 0.031 

0.330 ± 0.044 

0.330 ± 0.037 

0.37 % 

0.91 % 

2.37 % 

1.6 % 

1.96 % 

1.63 % 

2.17 % 

2.76 % 

0.66 % 

0.57 % 

1.32 % 

1.75 % 

2.31 % 

2.28 % 

0.82 % 

0.15 % 

0.337 ± 0.036 

0.337 ± 0.033 

0.342 ± 0.040 

0.342 ± 0.037 

-0.06 % 

-0.06 % 

0.330 ± 0.037 

0.332 ± 0.038 

0.340 ± 0.037 

0.350 ± 0.040 

0.59 % 

0.350 ± 0.037 

0.351 ± 0.035 

0.356 ± 0.040 

0.364 ± 0.045 

0.26 % 

POPT_left 

0.373 ± 0.036 

0.376 ± 0.033 

0.394 ± 0.046 

0.398 ± 0.042 

0.95 % 

POPT_right 

0.398 ± 0.039 

0.400 ± 0.036 

0.418 ± 0.051 

0.421 ± 0.050 

0.58 % 

SCP_left 

SCP_right 

0.348 ± 0.039 

0.350 ± 0.041 

0.364 ± 0.038 

0.370 ± 0.033 

0.71 % 

0.349 ± 0.032 

0.349 ± 0.032 

0.360 ± 0.039 

0.365 ± 0.035 

0.02 % 

SLF_III_left 

0.322 ± 0.037 

0.322 ± 0.035 

0.344 ± 0.063 

0.348 ± 0.061 

0.03 % 

SLF_III_right 

0.321 ± 0.038 

0.320 ± 0.038 

0.334 ± 0.055 

0.337 ± 0.054 

-0.2 % 

SLF_II_left 

0.298 ± 0.032 

0.299 ± 0.032 

0.309 ± 0.045 

0.316 ± 0.052 

0.13 % 

SLF_II_right 

0.298 ± 0.036 

0.297 ± 0.035 

0.309 ± 0.042 

0.314 ± 0.045 

-0.54 % 

SLF_I_left 

SLF_I_right 

STR_left 

STR_right 

0.283 ± 0.030 

0.284 ± 0.029 

0.291 ± 0.036 

0.294 ± 0.032 

0.45 % 

0.297 ± 0.033 

0.297 ± 0.031 

0.309 ± 0.043 

0.311 ± 0.040 

0.12 % 

0.416 ± 0.041 

0.423 ± 0.044 

0.453 ± 0.052 

0.467 ± 0.055 

1.72 % 

0.439 ± 0.034 

0.441 ± 0.036 

0.466 ± 0.057 

0.474 ± 0.056 

0.45 % 

ST_FO_left 

0.327 ± 0.039 

0.323 ± 0.039 

0.358 ± 0.053 

0.362 ± 0.050 

-1.24 % 

ST_FO_right 

0.351 ± 0.040 

0.347 ± 0.038 

0.387 ± 0.048 

0.397 ± 0.048 

-1.32 % 

ST_OCC_left 

0.341 ± 0.037 

0.343 ± 0.037 

0.353 ± 0.036 

0.363 ± 0.039 

0.48 % 

 322 

1.51 % 

1.69 % 

1.09 % 

2.22 % 

4.09 % 

0.44 % 

1.34 % 

2.06 % 

2.77 % 

2.01 % 

0.19 % 

1.61 % 

1.25 % 

2.85 % 

2.17 % 

1 % 

0.76 % 

1.61 % 

1.44 % 

1.08 % 

1.05 % 

2.52 % 

1.47 % 

1.04 % 

0.82 % 

3.07 % 

1.69 % 

1.27 % 

2.76 % 

2.75 % 

-0.342 

0.891 

1.237 

0.595 

-1.032 

2.296 

1.404 

-1.089 

-1.603 

-1.200 

-0.141 

0.033 

0.073 

1.065 

1.624 

-2.420 

-1.073 

-1.141 

-0.467 

-0.781 

-0.454 

-0.046 

0.714 

0.725 

0.995 

-0.807 

0.017 

0.041 

-0.417 

0.091 

-1.030 

-0.443 

-1.334 

-0.952 

-1.452 

-0.033 

-0.046 

0.451 

-0.282 

0.999 

-0.669 

-0.240 

-2.866 

-0.659 

1.326 

1.307 

0.62 

0.47 

0.47 

0.564 

0.47 

0.18 

0.47 

0.47 

0.47 

0.47 

0.682 

0.682 

0.682 

0.47 

0.47 

0.155 

0.47 

0.47 

0.582 

0.496 

0.582 

0.682 

0.518 

0.518 

0.47 

0.493 

0.682 

0.682 

0.582 

0.682 

0.47 

0.582 

0.47 

0.47 

0.47 

0.682 

0.682 

0.582 

0.637 

0.47 

0.532 

0.652 

0.107 

0.532 

0.47 

0.47 

-0.820 

0.493 

-1.239 

-1.748 

-0.742 

-2.107 

-1.365 

-0.171 

-1.129 

-1.614 

-2.255 

-1.399 

-0.219 

-2.011 

-1.346 

0.062 

-0.463 

-2.307 

-1.966 

-2.278 

-1.593 

-0.772 

-0.806 

-0.473 

-0.369 

-1.400 

-2.465 

-2.162 

-1.604 

-0.858 

-0.130 

0.073 

-3.376 

-3.475 

-1.387 

-1.540 

-1.661 

-1.394 

-0.640 

-0.896 

-2.782 

-2.063 

-0.698 

-0.661 

-2.847 

-1.822 

-0.608 

-2.003 

-3.437 

0.091 

0.059 

0.161 

0.052 

0.08 

0.258 

0.106 

0.067 

0.05* 

0.078 

0.251 

0.054 

0.08 

0.267 

0.202 

0.05* 

0.055 

0.05* 

0.067 

0.158 

0.154 

0.202 

0.22 

0.078 

0.048* 

0.052 

0.067 

0.146 

0.263 

0.267 

0.014* 

0.014* 

0.078 

0.07 

0.064 

0.078 

0.174 

0.142 

0.035* 

0.052 

0.167 

0.172 

0.035* 

0.056 

0.177 

0.054 

0.014* 

 
 
 
 
 
 
 
 
Table A.28 (cont’d)  

mTBI 

Controls 

TractSeg Name 

P1 

P2 

P1 

P2 

Percent 
Change 
mTBI % 

Percent 
Change 
Controls % 

t-score 
mTBI 

q-value 
mTBI1 

t-score 
controls 

q-value 
controls 

ST_OCC_right 

0.351 ± 0.036  0.352 ± 0.033  0.360 ± 0.041  0.367 ± 0.045 

0.19 % 

2 % 

-0.303 

0.634 

-3.482 

0.014* 

ST_PAR_left 

0.334 ± 0.032  0.337 ± 0.031  0.354 ± 0.043  0.358 ± 0.040 

0.84 % 

ST_PAR_right 

0.357 ± 0.035  0.357 ± 0.032  0.372 ± 0.045  0.375 ± 0.045 

0.25 % 

ST_POSTC_left 

0.330 ± 0.030  0.336 ± 0.033  0.364 ± 0.043  0.372 ± 0.041 

1.88 % 

1.01 % 

0.86 % 

2.18 % 

-1.248 

-0.431 

-2.931 

0.47 

0.582 

0.107 

-1.247 

-1.874 

0.091 

0.056 

-2.430 

0.048* 

ST_POSTC_righ
t 

0.357 ± 0.031  0.359 ± 0.031  0.385 ± 0.047  0.390 ± 0.047 

0.67 % 

1.27 % 

-1.208 

0.47 

-1.770 

0.059 

ST_PREC_left 

0.346 ± 0.029  0.352 ± 0.031  0.388 ± 0.039  0.399 ± 0.038 

1.79 % 

ST_PREC_right 

0.367 ± 0.030  0.370 ± 0.030  0.395 ± 0.043  0.402 ± 0.044 

0.73 % 

ST_PREF_left 

0.339 ± 0.030  0.338 ± 0.031  0.367 ± 0.039  0.372 ± 0.040 

-0.25 % 

ST_PREF_right 

0.340 ± 0.029  0.339 ± 0.028  0.367 ± 0.041  0.374 ± 0.043 

-0.56 % 

ST_PREM_left 

0.316 ± 0.030  0.318 ± 0.030  0.336 ± 0.044  0.344 ± 0.048 

0.5 % 

ST_PREM_right  0.331 ± 0.031  0.334 ± 0.032  0.357 ± 0.041  0.363 ± 0.040 

0.78 % 

T_OCC_left 

0.324 ± 0.036  0.326 ± 0.037  0.334 ± 0.035  0.344 ± 0.039 

0.54 % 

T_OCC_right 

0.342 ± 0.037  0.343 ± 0.034  0.349 ± 0.040  0.356 ± 0.045 

0.28 % 

T_PAR_left 

0.334 ± 0.032  0.336 ± 0.031  0.352 ± 0.042  0.356 ± 0.041 

0.75 % 

T_PAR_right 

0.361 ± 0.036  0.362 ± 0.033  0.374 ± 0.044  0.378 ± 0.045 

0.34 % 

T_POSTC_left 

0.347 ± 0.034  0.353 ± 0.036  0.377 ± 0.047  0.385 ± 0.048 

1.52 % 

T_POSTC_right 

0.368 ± 0.034  0.371 ± 0.034  0.395 ± 0.052  0.401 ± 0.052 

0.64 % 

T_PREC_left 

0.365 ± 0.033  0.371 ± 0.034  0.411 ± 0.041  0.422 ± 0.043 

1.71 % 

T_PREC_right 

0.382 ± 0.032  0.385 ± 0.032  0.411 ± 0.045  0.419 ± 0.046 

0.75 % 

T_PREF_left 

0.342 ± 0.031  0.342 ± 0.031  0.371 ± 0.039  0.377 ± 0.041 

-0.08 % 

T_PREF_right 

0.331 ± 0.028  0.330 ± 0.028  0.357 ± 0.039  0.364 ± 0.042 

-0.37 % 

T_PREM_left 

0.331 ± 0.031  0.333 ± 0.032  0.356 ± 0.043  0.363 ± 0.047 

0.57 % 

T_PREM_right 

0.344 ± 0.032  0.347 ± 0.034  0.373 ± 0.041  0.381 ± 0.042 

0.88 % 

UF_left 

0.311 ± 0.036  0.307 ± 0.035  0.334 ± 0.051  0.334 ± 0.047 

-1.09 % 

2.66 % 

1.82 % 

1.53 % 

1.96 % 

2.17 % 

1.85 % 

2.92 % 

2.03 % 

1.06 % 

0.98 % 

2.14 % 

1.47 % 

2.83 % 

2.02 % 

1.71 % 

1.97 % 

2.05 % 

2.23 % 

0.18 % 

-2.754 

0.107 

-2.424 

0.048* 

-1.210 

0.47 

-2.102 

0.052 

0.07 

0.056 

0.052 

0.07 

0.460 

1.296 

0.582 

-1.519 

0.47 

-1.897 

-0.803 

0.493 

-2.065 

-1.356 

-0.929 

0.47 

0.47 

-1.527 

-3.414 

0.014* 

-0.469 

0.582 

-3.460 

0.014* 

-1.119 

-0.553 

0.47 

0.58 

-1.497 

-1.920 

0.071 

0.056 

-2.419 

0.155 

-2.468 

0.048* 

-1.178 

0.47 

-1.840 

-2.717 

0.107 

-2.307 

-1.186 

0.47 

-2.159 

0.147 

0.890 

-0.954 

-1.481 

1.211 

0.682 

-1.740 

0.47 

0.47 

0.47 

0.47 

-1.941 

-1.711 

-1.829 

-0.062 

0.056 

0.05* 

0.052 

0.059 

0.055 

0.061 

0.056 

0.267 

0.308 ± 0.031  0.305 ± 0.031  0.335 ± 0.035  0.341 ± 0.040 

UF_right 
Tables A.28- T-test results for comparisons of the automatically segmented tracts from TractSeg 
between timepoints within mTBI and control subjects. 
1Significant values are marked with an asterisk (*) or <0.001 

-0.79 % 

1.88 % 

-0.942 

0.918 

0.47 

0.135 

 323 

 
 
 
 
 
 
 
 
 
Joshua H. Baker, M.S. 
3300 Auburn Rd. Ste 406 Auburn Hills MI, 48326 
610-442-4695 • bakerj77@msu.edu 

Education 
Doctor of Osteopathic Medicine 
2026 
COMLEX Level 1 & USMLE Step 1   
Michigan State University College of Osteopathic Medicine  
East Lansing, MI 

Doctor of Philosophy, Neuroscience 
2024 
College of Natural Sciences, Michigan State University 
East Lansing, MI 

Master of Science, Anatomy  
The Pennsylvania State University College of Medicine  
Hershey, PA 

Bachelor of Science, Psychology 
Minor in Chemistry 
The University of Pittsburgh 
Pittsburgh, PA 

Professional Positions and Experience 
ScholarRx Visual Designer    
Present 
Independent Contractor, Visual Bricks Project 
Remote 

Human Research Technologist II   
May 2018 
Department of Psychiatry, Sleep Research and Treatment Center 
The Pennsylvania State University College of Medicine  
Hershey, PA 

Adjunct Faculty 
May 2018 
Department of Physical Therapy 
Lebanon Valley College  
Annville, PA 

Expected May 

Passed 2021 

Expected May 

May 2017                

May 2015 

Jan 2023 to 

Jul 2017 to 

Jan 2017 to 

Certification 
Introduction to Interventional Radiology and Minimally Invasive Procedures     May 2023 
American Medical Student Association Scholars Program: “Racism in Medicine” 
ACLS/BCLS 

     Dec 2022 
Jun 2019 to Present 

 324 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Translational Science Certificate 
Conceptual Foundation of Medicine Certificate 

Research and Laboratory Experience 
Graduate Research Assistant 
Department of Radiology, Cognitive Imaging Research Center 
Michigan State University 
East Lansing, MI 

Human Research Technologist II   
Department of Psychiatry, Sleep Research and Treatment Center 
The Pennsylvania State University College of Medicine  
Hershey, PA 

Graduate Student, Master’s Thesis Project  
Department of Psychiatry, Sleep Research and Treatment Center 
The Pennsylvania State University College of Medicine  
Hershey, PA 

Undergraduate Research Assistant   
Western Psychiatric Institute and Clinic 
Pittsburgh, PA 

     May 2017 
     May 2015 

Aug 2018 to Present 

Jul 2017 to May 2018 

Jan 2016 to May 2017 

Nov 2011 to May 2015 

Professional Memberships  
The American Society of Functional Neuroradiology   
The International Society for Magnetic Resonance in Medicine  
American Osteopathic College of Radiology 
American College of Radiology  
American Society of Neuroradiology    
American Medical Student Association 
American Physician Scientist Association 

2024 to Present 

2024 to Present 
2022 to Present 
2022 to Present 

2022 to Present 

2019 to Present 
2018 to Present 

Software & Research Tool Competencies 
Analysis of Functional Neuroimages (AFNI)   
FMRIB Software Library (FSL)  
Freesurfer 
I2b2 
Linux   
MATLAB 
MRTrix3 
R & R Studio 
REDCap 
SLURM 
Statistical Parametric Mapping (SPM) 

Publications 

 325 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
             
 
 
Baker JH, Bender A, Scheel N, Zhu DC. A Rapid Review of  Signal Modeling Techniques for 
Diffusion Magnetic Resonance Imaging Beyond DTI: Findings in Mild Traumatic Brain Injury. In 
preparation. 

Baker JH, Bender A, Scheel N, Ding K, Zhu DC, The TRACK-TBI Investigators. A longitudinal 
Fixel-based analysis of mild traumatic brain injury patients: A TRACK TBI Study. In preparation. 

Baker JH, Schoppe K, Youmans D, Moriarity A. Scope of Practice Legislation Across the US: 
Current Trends in Evidence, Advocacy, and Action. Appl Radiol. 2023 Sep 12. 

Fernandez Z, Scheel N, Baker JH, Zhu DC. Functional connectivity of cortical resting-state 
networks is differentially affected by rest conditions. Brain Res. 2022 Dec 1;1796:148081. doi: 
10.1016/j.brainres.2022.148081. Epub 2022 Sep 12. PMID: 36100086. 

Baker JH. 2017. Insomnia Symptoms and Systemic Inflammation in Adolescents: A Population-
Based Study. The Pennsylvania State University College of Medicine, Hershey, PA. 
https://etda.libraries.psu.edu/catalog/13881jxb5933 

Fernandez-Mendoza J, Baker JH, Gaines J, Liao D, Vgontzas A, Bixler E. Insomnia Symptoms 
with Objective Short Sleep Duration is Associated with Systemic Inflammation in Adolescents. 
Brain Behavior Immunity. 2017. 61: 110-116.    

Baker JH, Rothenberger S, Kline C, Okun M (2016): Exercise During Early Pregnancy is 
Associated With Greater Sleep Continuity. Behavioral Sleep Medicine. 2016. 14:1-14.  

Posters/Oral Presentations 

Scheel N, Hubert J, Baker JH, Fernandez Z, Vongpatanasin W, Zhang R, Zhu DC. Comparison of 
T2 FLAIR WM Hyperintensity Segmentation Algorithms in Clinical Trials. Michigan Alzheimer’s 
Disease Research Center Beyond Amyloid Research Symposium. 2023. Poster Presentation. 

Baker JH, Bender A, Scheel N, Hubert J, Fernandez Z, Zhu, DC. A Pilot Longitudinal Diffusion-
Weighted Magnetic Resonance Imaging Fixel-Based Analysis in a Sample of Mild Traumatic Brain 
Injury Patients: A TRACK-TBI Study. Michigan State University College of Osteopathic Medicine 
Research Day. 2023. Poster Presentation 

Fernandez Z, Scheel N, Baker JH, Zhu DC. Functional connectivity of cortical resting-state 
networks is differentially affected by rest conditions. AAP/ASCI/APSA JM. Chicago, Illinois on 
April 21-23. Poster Presentation. 

Baker JH (host & presenter), Gadde JA (mentor & article author). American College of Radiology 
Medical Student Journal Club “Imaging of Hypoxic-Ischemic Injury in the Era of Cooling”. January 
6, 2023. Virtual National Oral Presentation. 

Baker JH, Brauman S, Tilden SE, Sadasivan M. Piloting Undergraduate Medical Student 
Participation in Curricular Change: A Lilly Fellowship Project. Michigan Osteopathic Association 
Autumn Scientific Research Exposition. 2022. Poster presentation.   

 326 

 
 
 
 
 
 
 
 
 
 
 
 
Fernandez Z, Baker JH, Scheel N, Zhu D. Functional Connectivity of Resting-State Networks is 
Affected by Excess Environmental Stimuli. Abstract submitted to the 26th Annual Meeting of the 
OHBM, June 26-30 2020, Montreal Canada. 

Baker JH, Vgontzas A, Fernandez-Mendoza J, Krishnamurthy B, Gaines J, Basta M, Criley C, 
Bixler EO. Effects of Trazodone on Blood Pressure: A Longitudinal, Observational Study of 
Patients Presenting to a Sleep Disorder Clinic. 32nd Annual Meeting of the Associated Professional 
Sleep Societies. Baltimore, Maryland on June 3-6, 2018. Associated Professional Sleep 
Societies. Poster presentation. SLEEP 2018; 41 (Abstract Supplement): A157. 

Vgontzas A, Fernandez-Mendoza J, Baker JH, Krishnamurthy B, Gaines J, Calhoun S, Basta M, 
Bixler EO. Trazodone vs. Cognitive Behavioral Therapy in Insomnia with Short Sleep Duration: Effects on Total 
Sleep Time and Cortisol Levels. 32nd Annual Meeting of the Associated Professional Sleep Societies. 
Baltimore, Maryland on June 3-6, 2018. Associated Professional Sleep Societies. Oral 
presentation. SLEEP 2018; 41 (Abstract Supplement): A142-A143. 

Fernandez-Mendoza J, Gaines J, Baker JH, Calhoun S, Liao D, Bixler E, Vgontzas A. Insomnia is 
Associated with Increased C-reactive Protein levels in Adolescents: Role of Objective Short Sleep 
Duration. 23rd Congress of the European Sleep Research Society. Bologna, Italy, September 13-16 
2016. European Sleep Research Society. Oral presentation. J Sleep Res. 2016; (abstract supplement). 

Okun M, Baker JH, Rothenberger S, Kline C The Effect of Exercise on Sleep During Pregnancy 
Paternal Maternal Stress and Child Outcome: Exploring Novel Underlying Mechanisms Symposium. 
Society for Research in Child Development Oral Presentation. Philadelphia, Pennsylvania SRCD 
2015  

Awards and Scholarships 

2024 Medical Student Scholarship    
American College of Radiology 
Reston, VA 

Dissertation Completion Fellowship  
College of Natural Sciences, Michigan State University  
East Lansing, MI 

Second Place Poster Competition Award Winner   
Michigan State University College of Osteopathic Medicine Research Day 
Novi, MI 

Graduate Teaching Assistant Scholarship   
The Pennsylvania State University College of Medicine 
Hershey, PA 

Peer-Review 

Jan 2024 

May 2023 

May 2023 

May 2016 

 327 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Early Career Section of Editorial Advisory Board   
Applied Radiology 

Peer-Reviewer 
Spartan Medical Journal  

Peer-Reviewer 
ScholarRx Bricks Create Grant 

Peer-Reviewer, supervised by Dr. David Zhu  
British Journal of Sports Medicine 

Peer-Reviewer, supervised by Dr. Fernandez-Mendoza 
Pediatrics 

Peer-Reviewer, supervised by Dr. Fernandez-Mendoza 
Brain Behavior and Immunity 

Peer-Reviewer, supervised by Dr. Fernandez-Mendoza 
2017 
Journal of Sleep Research 

Educational Activities 

ScholarRx Visual Designer   
2024 
Independent Contractor, Video Bricks Project 

-“The Electrocardiogram Leads and Localization” 
-“Cardiac Cycle: Pressure-Volume Loop” 
-“Pathophysiology of Aortic Stenosis” 

Lilly Fellowship Contributor 
Michigan State University  

-Autonomic control of bronchial smooth muscle 
-Introduction to the arachidonic acid pathway 
-Introduction to catecholamine synthesis 

Adjunct Faculty, Department of Physical Therapy 
Lebanon Valley College  

-Physiology Laboratory Instructor 
-Kinesiology Laboratory Instructor 
-Exercise Physiology Laboratory Instructor  

   Aug 2023 to Present 

     Jan 2022 to Present 

        May 2022 

        Mar 2019 

         Mar 2017 

          Jan 2017 

          Oct 

           March 2023 to Jan 

Aug 2021 to May 2022 

Jan 2017 to May 2018 

Graduate Teaching Assistant, MSDR Block  
The Pennsylvania State University College of Medicine 
-Cadaveric Dissection Lab Instructor   
-Lecture, Anatomy of the Vertebral Column and Spinal Cord   

Oct 2016 to Dec 2016                                                                      

 328 

 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
                                                                                                                            
Graduate Teaching Assistant, Neurology Residency Program 
The Pennsylvania State University College of Medicine  
-Lecture, Cerebellum and Cerebellar Vasculature 
-Lecture, Cerebral Vasculature  

Graduate Teaching Assistant, GI/GU Block 
The Pennsylvania State University College of Medicine 
-Cadaveric Dissection Lab Instructor   
-Lecture, Anatomy of the Retroperitoneum 

Graduate Teaching Assistant, Physician Assistant Program  
The Pennsylvania State University College of Medicine  
-Cadaveric Prosection Lab Instructor   
-Lecture, Anatomy of the Pelvis  

Social Media  
Twitter: @DO_Radiology 
Instagram: @do_radiology 

Service & Leadership activities 

ACR Medical Student Journal Club Moderator  
American College of Radiology 

Michigan Radiologic Society Class of 2026 School Liaison  
Michigan Radiologic Society 

Education Lead, Medical Student Section   
American College of Radiology 

Co-Founder and Member 
MSUCOM Radiology Interest Group 
Michigan State University College of Osteopathic Medicine  

Student Mentor 
Summer Research Opportunity Program 
Michigan State University 

Rad Reserves Volunteer 
The Radiology Room 
American College of Radiology 

Council Member 
Student Advisory Council 
ScholarRx, First Aid for the USMLE  

 329 

        Oct 2016           

       Aug 2016                                                     

May 2016 to Sep 2016                                                                                  

May 2023 to Present 

 Sept 2023 to Present 

 May 2023 to Present 

   Jan 2023 to Present 

       Sept 2022 

  July 2022 to Present 

  Aug 2021 to Present 

 
             
 
 
 
 
 
 
 
 
 
        
 
             
 
           
 
 
 
 
 
 
 
 
 
          
 
 
 
 
 
        
 
 
 
   
 
 
 
 
 
 
 
           
 
 
 
 
 
 
 
 
 
 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Committee Member   
DO/PhD Student Advisory Committee 
Michigan State University DO/PhD Program  

Volunteer Lecturer 
Brain Awareness Week 
Metro Detroit Public Schools 

Volunteer Instructor   
Present 
Physiology PhUn Day 
Impression 5 Science Center, Lansing MI 

Co-Founder and Member 
Society for Applied Anatomy 
The Pennsylvania State University College of Medicine 

Secretary and Member 
Pathology Interest Group 
The Pennsylvania State University College of Medicine 

Volunteer 
Ronald McDonald House 
The Pennsylvania State University College of Medicine  

May 2019 to Apr 2020 

    Mar 2019 to Present 

     Oct 2018 to 

Aug 2016 to May 2017 

Aug 2016 to May 2017 

Nov 2016 to May 2018 

 330 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Volunteer Lab Instructor 
Healthcare Career Exploration Pre-College Program 
The Pennsylvania State University College of Medicine  

Volunteer Lab Instructor 
2016  
Central Pennsylvania Brain Bee 
The Pennsylvania State University College of Medicine 

Volunteer Lab Instructor 
2015  
Central Pennsylvania Brain Bee 
The Pennsylvania State University College of Medicine 

Aug 2016 to May 2018 

          Feb 

          Feb 

 331