440 Deep Learning Enables Automated Evaluation Of Global Longitudinal Shortening From Cine CT

Abstract

Introduction: CineCT is increasingly being used to evaluate cardiac dynamics. However, current evaluation typically depends on visual assessment, 3D segmentation, or evaluation based on wall thickening. Echocardiography and CMR have demonstrated the utility of global longitudinal shortening (GLS). Currently measuring GLS with CT requires reformatting the 4D dataset into long-axis imaging planes and delineating the endocardial boundary across time. In this work, we demonstrate the ability of a deep learning framework to automatically and accurately measure GLS for detection of wall motion abnormalities (WMA).