Cascaded Convolutional Neural Network for Automatic Myocardial Infarction Segmentation from Delayed-Enhancement Cardiac MRI

Automatic segmentation of myocardial contours and relevant areas like infraction no-reflow is an important step for the quantitative evaluation infarction. In this work, we propose a cascaded convolutional neural network automatic infarction from delayed-enhancement cardiac MRI. We first use 2D U-Net to focus on intra-slice information perform preliminary segmentation. After that, 3D utilize volumetric spatial subtle Our method evaluated MICCAI 2020 EMIDEC challenge dataset achieves average Dice score 0.8786, 0.7124 0.7851 myocardium, respectively, outperforms all other teams contest.