Computational Fluid Dynamics in Highly Complex Geometries Using MPI-Parallel Lattice Boltzmann Methods: A Biomedical Engineering Application

Abstract

This study aims to establish predictive criteria for identifying cerebral aneurysms that are likely to respond favorably--either stabilizing or shrinking--following flow diverter stent (FDS) treatment. We analyzed the pre-treatment hemodynamics and geometry of four patient-specific aneurysms to determine features linked to positive outcomes. Blood flow was simulated using a massively parallel, in-house developed CFD code. Hemodynamic metrics--including vortex structure, velocity field, wall shear stress (WSS), time-averaged WSS (TAWSS), and oscillatory shear index (OSI)--were quantified in each case. Aneurysms that responded well to FDS showed lower OSI and elevated WSS and TAWSS near the neck region. In contrast, poor responders exhibited larger vortexes and persistently low WSS and TAWSS within the sac, accompanied by high OSI. Geometric factors, such as smaller neck and sac sizes and greater distance from the skull base, also appeared to favor a positive response. These results highlight a combination of geometric and hemodynamic parameters that may serve as effective predictors of FDS treatment success and support more informed decision-making during preoperative planning.