Automatic classification and segmentation of atherosclerotic plaques in the intravascular optical coherence tomography (IVOCT)

•Propose a framework for the plaque classification and pixel-level segmentation in IVOCT images.•Use the prior that plaques are often distributed in the wall near the outline of the lumen to design a patch cropping scheme for the classification and a polar transform scheme for the segmentation to information-lossless reduce the image size.•The accuracy of the plaque classification is close to 1.