DOI: 10.11648/j.ijbecs.20261202.12 ISSN: 2472-1301

Pectoral Muscle Detection and Removal in Mammograms Using Information Gain Based FCM Algorithm

Navneet Kaur
Show PDF Cite
The widest used modality for detection of breast cancer is Mammography. Automated cancer detection is a tedious work because there are many types of high intensity artifacts found in MLO views of mammograms. So, preprocessing here becomes an essential step for successful classification. To get the region of interest, we carried out all the steps including removal of pectoral muscle as it has high intensity level similar to the intensity level of abnormalities already present in the mammogram so if we will take the input image as it is including the pectoral muscle and other artifacts it will give inaccurate results. This paper proposes a methodology for segmenting and removing the pectoral muscles using a technique named information gain based fcm. The proposed algorithm combines fuzzy C mean and neutroscopic L-means methods for correctly extracting the pectoral muscle. To validate the work, the proposed methodology is tested on different mammographic images of MIAS database and the results obtained nearly follows that marked by an expert radiologist. The performance of the proposed approach is evaluated using several quantitative metrics, including Hausdorff Distance (HD), Mean Error (ME1), Relative Foreground Area Error (RFAE), Misclassification Rate (ME2), Extraction Error Rate (EER), and Region Non-Uniformity (RNU).

More from our Archive