Abstract
This study presents a computational framework to perform CT-based, personalized blood flow analysis in human left atrium (LA). Patient-specific LA geometry and function were obtained from the cardiac CT images of a patient acquired during sinus rhythm. A four-dimensional displacement vector field was estimated using a non-rigid registration method. The LA blood outflow across the mitral valve was calculated from LV volume change, and the flow field within the LA was derived from computational fluid dynamics (CFD) using the incompressible Navier-Stokes equation. The global characteristics of the LA motion and blood flow excellently agreed with clinical observations. The personalized blood flow analysis in the LA may be useful to evaluate the flow stasis associated with the risk of the intracardiac thrombosis and stroke.
