Title:  Machine Learning Approaches to Predicting Postoperative Hemodynamics after Cardiac Surgeries

Committee: 

Dr. Dasi, Advisor

Dr. AlRegib, Co-Advisor

Dr. Anderson, Chair

Dr. Oshinski

Abstract: The object of the proposed research is to develop machine learning approaches to preoperatively predicting postoperative hemodynamics for cardiac surgeries. The initial effort explores to predict postoperative boundary conditions in Fontan procedures to shift the current Fontan surgical planning paradigm where preoperative boundary conditions have been inaccurately used towards the simulation-based selection of surgical options. Based on the machine learning approach developed in Fontan surgical planning, this research seeks to further propose a machine learning rationale for the postoperative hemodynamics prediction required for the more accurate surgical planning or diagnosis within a wider range of cardiac surgeries including heart valve replacements.