Project Summary/Abstract Tetralogy of Fallot (ToF) is the most common cyanotic congenital heart disease, affecting 0.3% of children. Before correction, its four defects lead to increased right heart pressures and mixing of oxygenated and deoxygenated blood. Even after surgical repair, patients may experience elevated right heart pressures and volumes due to residual pulmonary stenosis, pulmonary regurgitation, scar formation, and conduction abnormalities. These changes in geometry, wall thickness, and pressure-volume relationships all contribute to right ventricular (RV) remodeling, which can eventually lead to adverse events such as ventricular arrhythmias, RV dysfunction, and the need for pulmonic valve repair, affecting up to 44% of patients overall. Despite the great advances that have been made in medical and surgical care of ToF patients, there is still limited understanding of which patients will experience adverse RV remodeling and subsequent clinical events. ToF patients’ cardiac function is normally assessed annually using cardiovascular magnetic resonance (CMR) imaging, which provides excellent views of the right heart and its valves. However, manual analysis of these images is time-consuming and subject to inter- and intra-user variability. Additionally, CMR provides anatomic and flow data enabling quantification of pulmonary artery hemodynamics, but has not yet been investigated in post-repair ToF patients. Detailed characterization of pulmonary artery stresses and pressures, through the application of computational fluid dynamics (CFD) simulations, could provide insight into factors affecting RV remodeling. There remains an unmet need to comprehensively identify features that characterize and predict progression from primary ToF repair to adverse RV remodeling and poor outcomes. My objectives in this proposal are to identify the structural and hemodynamic parameters of ToF that are associated with RV remodeling in order to improve both clinical care and quality of life. I plan to approach these objectives using two specific aims. In Aim 1, I will develop a supervised machine learning algorithm to accurately and automatically segment 3D cardiac volumes using CMR images. This algorithm will enable robust and repeatable measurements of cardiac structure and function for both cross-sectional and longitudinal analyses. I hypothesize that this algorithm will achieve accurate and precise segmentation results as assessed by Dice scores and intraclass correlation coefficients. In Aim 2, I will study patient-specific pulmonary artery hemodynamics and determine associations with adverse RV remodeling. Specifically, I will generate 3D and 1D CFD models based upon CMR-derived geometries and phase-contrast flow data. I hypothesize that hemodynamic parameters such as wall shear stress and total pathway resistance will be associated with and provide mechanistic insight into RV remodeling. Overall, I anticipate that this project will provide me with...