Automatic Enlarged Lymph Node Detection from 3D Mediastinum CT Images

Lymph nodes (LNs), part of the lymphatic system, have important role of the immune system. also have an important function in current clinical workflow. LN routinely assessed by clinicians cause of related to all kinds of cancer, cancer affected the lymph nodes to be enlarged, lymph nodes nearby primary cancer regions, accurate lymph node detection very important for cancer staging and to monitor disease progress also to prediction surgical planning, and to follow-up effectiveness of treatment and therapy. The detection and assessment of lymph nodes is usually performed manually, since finding the lymph nodes can be time – consuming and depend on the highly observer experience during the low contrast of surrounding structures in computed tomography (CT), a system for lymph node detection desired, so i proposed a method for automatic lymph node detection and measurement from 3D Mediastinum CT images by employ’s a combination between the speeded up robust features (SURF) and the blob detection, to produce a system with high accuracy using feature extraction method, Also produce algorithm with low computational time by analyzing multi slices by converted to one projected image each time using maximum intensity projection, also to eliminating vessels, and add new calculation parameter to the system which is the volume of the LN, to add extra accuracy for the proposed method, which allow the system from verifying the candidate LN., the method is evaluated for Mediastinum LN detection on 50 volumes containing 182 LNs, yielding an 83.3%, detection rate with 1.1 false positive per volume, with an average running time of 12-35 seconds per volume.