Comparison of algorithms for 3D arteries segmentations on WBMRA images

Different methods have been proposed in literature to solve the problem of segmentation of vascular structures. Here, we present a comparison of four different solutions: the level set method and three classifiers that use hand crafted features, two based on simple threshold and one on SVM. We first take a look at the theory behind each techniques then we explain how we implemented them in our work in order to create an au- tomatic segmentation method that can be applied inter-patient.