| dc.description.abstract | Abstract. Computational fluid dynamics (CFD) is commonly used to investigate hemodynamics in the cardiovascular system, particularly in regions prone to cardiovascular disease, such as the carotid artery bifurcation. Despite its potential, significant variability exists across different computational approaches, highlighting the need for systematic solver comparisons. This study provides a comprehensive evaluation of three open-source finite element method (FEM) solvers—SimVascular, FEBio, and FEniCS Oasis—for simulating blood flow in a subject-specific carotid artery model. We conducted a rigorous comparison using a model derived from 4D phase-contrast magnetic resonance imaging (4D Flow MRI), examining solver performance across multiple mesh resolutions. This analysis focused on key hemodynamic metrics, including velocity fields, time-averaged wall shear stress (TAWSS), oscillatory shear index (OSI), and wall shear stress (WSS) topology. By maintaining identical meshes, boundary conditions, and postprocessing methods, we isolated solver-specific characteristics while focusing on high-resolution mesh refinements. All solvers demonstrated similar capability in representing the 4D-Flow MRI data. Notably, all solvers consistently identified critical hemodynamic regions, such as flow disturbance zones in the carotid sinus. Mesh convergence analysis showed the ability of all solvers to achieve converged predictions at relatively lower mesh resolutions. The computational time was also compared across the three solvers. While demonstrating the capabilities of each solver in predicting physiologically relevant hemodynamic patterns, our study underscores the utility of open-source solvers for high-fidelity hemodynamic predictions. | |
