Large-Scale Lattice Boltzmann Method Simulations of Stenosed Carotid Bifurcation

Carotid artery diseases are extremely harmful to the human brain, especially unstable plaques. Wall shear stress is significant for clinical diagnosis of carotid plaque stability. The challenge of the traditional simulation method is that preprocessing time and computation time of large-scale simulation are relatively long. In this paper, a fast simulation procedure for two-dimensional (2D) and three-dimensional (3D) blood flow models of stenosed carotid based on parallel lattice Boltzmann method was proposed. The geometry model of carotid artery stenosis was established by medical image reconstruction method. For computational grid generation, in the 3D model, we used the wall encryption grid method to improve the precision of the near wall blood flow simulation and reduce the number of computational grids. The simulation results were compared with the analytic results to verify the rationality and accuracy of the model. In order to meet the clinical requirements, especially the limited diagnostic time, parallel optimization was carried out in high performance computing cluster. The simulation results show that the parallel algorithm with OpenMP libraries were more than 15 times faster than the serial simulations. Our method can assist doctors to diagnose carotid artery disease in a short time.