A fully automatic and multi-structural segmentation of the left ventricle and the myocardium on highly heterogeneous 2D echocardiographic data

Abstract

Although 3D ultrasound plays an increasingly important role, 2D echocardiography remains the main clinical imaging modality for cardiac function assessment in daily practice. This requires precise delineation of the myocardium at end diastole (ED) and systole (ES). Because of intrinsic high variability in image quality, manual interactions are still needed. In this study, we investigate a machine learning solution to fully automate the segmentation of the myocardium on heterogeneous dataset.