Automatic Detection and Segmentation of Lymph Nodes Using NonMaximal Suppression Algorithm from CT Data

Image segmentation is a difficult task in all medical imaging process. The major issue deals with the segmentation time needed for each organ. This paper presents a robust learning-based method for automatic detection and segmentation of solid lymph nodes from CT data.Lymph node is an oval-shaped organ of the immune system, which is difficult to identify in most cases and is distributed widely throughout the body so that they are having high clinical significance. The proposed approach first uses a classifier to train the lymph nodes to extract haar and self aligning features. Second, it presents a Nonmaximal suppression algorithm for the lymph node detection inorder to identify the most relevant lymph nodes. Third, it uses a segmentation based on Delaunay triangulation to segment the detected lymph nodes. The method is evaluated for abdominal and pelvic LN detection containing 371 LN, yielding a 84.0% detection rate with 1.0 false positive per volume.