Non-calcified plaque-based coronary stenosis grading in contrast enhanced CT

Segmentation of coronary artery tree from 3D cardiac CT scan followed by vessel stenosis grading using Deep learning. Restrospective CT image data was obtained from a public resource and active contour based segmentation was performed in the first step. The segmented coronary tree was evaluated by reconstructing 2D orthogonal slices from 3D space. Orthogonal slice-based individual segment was passed to CNN model for detection and classification of stenosis in respective segments.