Towards Physiologically Realistic Vascular Fluid-Structure Interaction Analysis Based on Medical Images
Please login to view abstract download link
Accurate biomechanical modeling is essential for assessing vascular diseases and guiding clinical decision-making. This requires techniques to reconstruct patient-specific vascular geometry with physiological details and generate high-quality meshes for fluid-structure interaction (FSI) analysis. In this work, we present a systematic pipeline for image-based vascular model generation, designed to address the lack of physiologically realistic modeling techniques. The pipeline enables the creation of thick-walled artery models and multi-layered tissue structures, capturing the distinct roles of the intima, media, and adventitia. Using a morphology-based strategy, we define local basis vectors, providing the foundation for describing tissue properties such as fiber orientation. Boolean operations further allow the integration of intraluminal thrombus (ILT), calcification, or atherosclerotic plaques into the vascular models. Coupled with a monolithic FSI solver based on the unified continuum formulation, this framework facilitates the analysis of hemodynamics and tissue biomechanics under physiological conditions. Two examples will be used to demonstrate the utility of the proposed methodology in clinical settings. First, isotropic and anisotropic material models are compared in FSI simulations, revealing significant differences in stress distributions. Building upon these results, multi-layered tissue models are integrated to investigate the dual role of ILT in abdominal aortic aneurysm (AAA) risk. While ILT reduces wall stress in covered areas, potentially mitigating rupture risk, it also promotes tissue degradation, elevating rupture risk. This study underscores the importance of physiologically detailed models for analyzing critical vascular biomechanical factors. It provides a scalable framework for studying AAA biomechanics and advancing personalized vascular simulations.