Person:

Feldman, Charles Lawrence

We present a computational method for commodity hardware-based clinical cardiovascular diagnosis based on accurate simulation of cardiovascular blood flow. Our approach leverages the flexibility of the Lattice Boltzmann method to implementation on high-performance, commodity hardware, such as Graphical Processing Units. We developed the procedure for the analysis of real-life cardiovascular blood flow case studies, namely, anatomic data acquisition, geometry and mesh generation, flow simulation and data analysis and visualization. We demonstrate the usefulness of our computational tool through a set of large-scale simulations of the flow patterns associated with the arterial tree of a patient which involves two hundred million computational cells. The simulations show evidence of a very rich and heterogeneous endothelial shear stress pattern (ESS), a quantity of recognized key relevance to the localization and progression of major cardiovascular diseases, such as atherosclerosis, and set the stage for future studies involving pulsatile flows.