Title: Computational Hemodynamics: Congenital and Acquired Heart Disease
Speaker: Steven Frankel
Date & Location: 4:00pm 3/24/2010 Lily Hall 3118
Abstract:
Cardiovascular disease is the single leading cause of death in America today. Estimates from the AHA for the year 2006 are that 81,100,000 people in the United States have one or more forms of cardiovascular disease (CVD). In addition, and no less important, thousands of infants born each year have congenital cardiovascular defects. Advances in mathematical models, numerical methods, medical imaging, and computational power have increased the role of computational hemodynamics in increasing insight into complex biofluid flows and aiding in the design and testing of medical devices.
In this talk, the primary focus is on the treatment of single functional ventricle as the leading cause of death from any birth defect in the first year of life. Currently, repair is performed in a complex series of 3 staged operations which are notorious for instability and mortality. The endpoint of repair is a univentricular Fontan circulation, in which the vena cavae are connected to the pulmonary artery. Because there is no right-sided ventricular power source, venous return is profoundly altered and preload to the single ventricle is suboptimal. We hypothesize that a means of augmenting existing Fontan cavopulmonary flow through the lungs would reverse these problems and produce a more stable two-ventricle physiology. We report computational results based on a breakthrough innovation, a so-called viscous impeller pump, which potentially solves this significant medical challenge.
The secondary focus of this talk is on the use of high-fidelity computer simulations to provide insight into the often complex cardiovascular flow physics. Specifically, we consider the Fontan circulation described above and simulate the flow through the so-called total cavopulmonary connection using high-order spectral-element and finite-difference based CFD approaches. These approaches are also applied to simulate steady and pulsatile flows in idealized stenotic blood vessels which is related to the problem of atherosclerosis. Finally, extensions of these approaches to more realistic (patient-specific) vessel geometries and accompanying flows is discussed.