Three-dimensional patient-specific models of abdominal aortic aneurysms (AAAs) were reconstructed from CT scans and vessel migration, eccentricity, and cross-section line curvature were identified as morphological features which are influenced by surrounding tissues during disease progression. A novel centerline algorithm was developed in Matlab and applied capable of qualitatively estimating vessel migration away from the vertebrae during disease progression and automatically parameterizing patient-specific surface models with respect to the generated centerline. The parameterized surface is then approximated using shape functions and reconstructed appropriately. Image analysis and reconstruction was performed through a commercial biomedical imaging software suite for a high-resolution CT dataset consisting of 51 scans across 11 patients which were obtained through collaboration with Dr. Whal Lee at Seoul National University Hospital (SNUH). A longitudinal dataset, multiple scans taken at multiple times per patient, presents additional challenges such as the need for accurate model registration and the unique opportunity of increasing our understanding of geometrical evolution during the progression of AAA. This study discovers that a rich and complex biomechanical relationship exists between an AAA and the spine during AAA disease progression. The most significant implication of these results highlights the necessity of incorporating surrounding tissue, whether patient-specific or idealized, for computational biomechanics stress simulations of AAAs to improve their accuracy and prediction of rupture potential. Incorporating such interactions would also considerably enrich documented analyses.
Read
