An Integrated Framework for Automated Image Segmentation and Personalized Wall Stress Estimation of Abdominal Aortic Aneurysms

Abdominal Aortic Aneurysm (AAA) remains a significant public health challenge, with an 82.1% increase in related fatalitiesfrom 1990 to 2019. In the United States alone, AAA complications resulted in an estimated 13,640 deaths between 2018and 2021. In clinical practice, computed tomography angiography (CTA) is the primary imaging modality for monitoring andpre-surgical planning of AAA patients. CTA provides high-resolution vascular imaging, enabling detailed assessments ofaneurysm morphology and informing critical clinical decisions. However, manual segmentation of CTA images is labor intensiveand time consuming, underscoring the need for automated segmentation algorithms, particularly when feature extractionfrom clinical images can inform treatment decisions. We propose a framework to automatically segment the outer wall of theabdominal aorta from CTA images and estimate AAA wall stress. Our approach employs a patch-based dilated modified U-Netmodel to accurately delineate the outer wall boundary of AAAs and and Nonlinear Elastic Membrane Analysis (NEMA) toestimate their wall stress. We further integrate Non-Uniform Rational B-Splines (NURBS) to refine the segmentation. Duringprediction, our deep learning architecture requires 17 ± 0.02 milliseconds per frame to generate the final segmented output.The latter is used to provide critical insight into the biomechanical state of stress of an AAA. This modeling strategy mergesadvanced deep learning architecture, the precision of NURBS, and the advantages of NEMA to deliver a robust, accurate, andefficient method for computational analysis of AAAs.