No abstract

No abstract

Detection of significant biomarker has been a major issue in the pathological study of cancerous tissue. The microscopic observation of H&E stained histopathological slides involves the study of various tissue objects and their significant traits influencing the cancer grading. The change in shape and size of the nuclei objects (nuclear pleomorphism) contributes much significantly in grading the cancer. With the advent of various imaging studies over a digitized Histopathological image, Active contour based segmentation approaches are considered to be much potential scheme in detecting the nuclei within occlusions and extracting their irregular boundaries. In this research, a novel approach of driving the contours of distance regularized level sets using an improved watershed transformation, has been presented. An Expectation Maximization based morphologically precomputed shape prior is used to extract the foreground markers, which controls the watershed transformation. The result of watershed transformation is used compute the centroid of nuclei to serve the initialization of the contour and the proposed gradient to drive the same efficiently. The study performed over the Benign and Malignant tissue images from BreakHis dataset has shown the efficacy of the methodology in terms of object detection accuracy and overlap resolution. The segmentation accuracy is compared to that of Geodesic active contours, based on the ground truth generated by expert pathologist.

No abstract

Nuclear pleomorphism is considered to be one of the most significant shape based feature adapted in grading the cancer through the pathological studies of the H&E stained tissue slides. Microscopic study of manually extracting the feature is highly laborious and misleads the pathologists during grading. Digitization of the slides has given rise to various segmentation approaches to extract the nuclei shape to assess the degree of pleomorphism. Here, a novel approach of initializing and evolving the distance regularized level sets (DRLS) for the detection and segmentation of the nuclei has been presented. In this work, two major objectives have been achieved. First, a novel geometric approach has been devised for the detection of centroids of each nuclei in the occluded region and second, a shape prior model has been presented for the extraction of gradient information through morphological operations. The multiple level set implementation of the DRLS contours are initialized using the centroids detected and driven through the gradient computed. The proposed method has been experimented over the images of benign and malignant breast cancer tissue obtained from BeakHis dataset. A quantitative analysis of the results have shown that a 97% of object detection accuracy and 78% of overlap resolution has been achieved through the proposed model. A comparative study with that of geodesic active contours have indicated an improvement in the segmentation accuracy measure of 9-10 pixel difference.