Leveraging Multiscale Blood Flow Modeling for Medical Device Therapy

Increasing prevalence of cardiovascular diseases such as heart failure have led to growing dependence on medical devices such as left ventricular assist devices (LVAD) to maintain adequate circulation for a patient with a failing heart. However, hemodynamically-induced complications such as emboli formation and thrombosis can lead to devastating adverse events such as pump failure and stroke. Blood as a fluid is truly unique, comprising of red blood cells, white blood cells and platelets suspended in plasma. Thus, the diverse and complex behavior of blood as a particle-laden fluid operating under various spatio-temporal interdependencies presents a fascinating and challenging problem to analyze. It is crucial to elucidate the complex interplay between hemodynamic stimuli and biological response, especially in the setting of blood-contacting medical devices such as LVADs. Several surgical and anatomical features such as left ventricle size, shape, LVAD implantation configuration, native contractility, aortic root structure and LVAD operational parameters have a significant influence on hemodynamic performance and efficacy of LVAD therapy. In this seminar, I will discuss some of the approaches we use at various scales (micro and macro) to analyze hemodynamics and associated complications such as stroke risk. Specifically, I will describe our recent methodologies involving virtual surgery, surgical optimization and designing patient-specific device management strategies to optimize hemodynamic performance. Further, we will also discuss similar approaches that can be utilized to investigate hemodynamic optimization neurovascular therapy such as cerebral aneurysm treatment. Finally, we will discuss some of the approaches that can be utilized in the future, such as combining machine learning with hemodynamic quantification for patient-specific hemodynamic optimization towards improving patient outcomes.