Advancing patient-specific models of blood flow for clinical decision support
Columbia University
Abstract
Computational models of blood flow have been developed to leverage their predictive capabilities and have been instrumental in elucidating mechanisms underlying flow-mediated disease progression, evaluating alternative surgical designs, and guiding treatment planning. However, limited success was achieved in translating these advances into clinical workflow. This is likely due to (i) data scarcity, which often leads to model generalization and reduces its patient-specificity, (ii) not accounting for the uncertainty in the model input, especially when the stakes are high, or (iii) the inadequacy of the methods in capturing the multifactorial nature of the disease or the therapy, including disease complexity, multiphysics interactions, and regulation and remodeling. In this talk, I will focus on two applications of predictive blood flow models of varying fidelity, depending on criticality, to guide clinical management and treatment planning. First, we will show how simple reduced–order models of whole-body circulation can be leveraged to assist pediatric heart surgeons in making informed, confident decisions for neonates born with undersized hearts. Second, we will employ complex fluid–structure interaction models of blood flow in the ascending aorta to analyze blood flow through the repaired aorta in patients with aneurysms, with the ultimate goal of designing individualized surgical grafts taking morphology and regional biomechanical interactions into account. An overview of other ongoing research projects will also be presented.
About the speaker
Dr. Vijay Vedula is an Assistant Professor in the Mechanical Engineering department at Columbia University, where he directs the Cardiovascular Biomechanics Research Lab (CBRL). He began his academic training in India, earned a Ph.D. in Mechanical Engineering from Johns Hopkins University, and underwent postdoctoral training at UCSD and Stanford. Dr. Vedula’s research is highly interdisciplinary, spanning cardiovascular engineering, computational biomechanics, 3M modeling (multiscale, multiphysics, and multifidelity), and, more recently, inverse problems and data-driven modeling for personalization. He is a recipient of multiple awards and fellowships, including a postdoctoral fellowship from the Child Health Research Institute at Stanford University, a von Kármán visitor fellowship at RWTH Aachen University, an Early Faculty Independence Award from the American Heart Association (AHA SCEFIA), and, more recently, the NSF CAREER award.
