We present a rectangle-based segmentation algorithm that sets up a graph and performs a graph cut to separate an object\r\nfrom the background. However, graph-based algorithms distribute the graph�s nodes uniformly and equidistantly on the\r\nimage. Then, a smoothness term is added to force the cut to prefer a particular shape. This strategy does not allow the cut\r\nto prefer a certain structure, especially when areas of the object are indistinguishable from the background. We solve this\r\nproblem by referring to a rectangle shape of the object when sampling the graph nodes, i.e., the nodes are distributed nonuniformly\r\nand non-equidistantly on the image. This strategy can be useful, when areas of the object are indistinguishable\r\nfrom the background. For evaluation, we focus on vertebrae images from Magnetic Resonance Imaging (MRI) datasets to\r\nsupport the time consuming manual slice-by-slice segmentation performed by physicians. The ground truth of the\r\nvertebrae boundaries were manually extracted by two clinical experts (neurological surgeons) with several years of\r\nexperience in spine surgery and afterwards compared with the automatic segmentation results of the proposed scheme\r\nyielding an average Dice Similarity Coefficient (DSC) of 90.9762.2%.
Loading....