Detection and elimination of pectoral muscle in mammogram images using Rough Set Theory

The pectoral muscle represents a predominant density region in most Medio-Lateral Oblique (MLO) view of mammograms. However, the presence of artifacts and pectoral muscle can disturb the detection of breast cancer and reduce the rate of accuracy in the Computer Aided Diagnosis (CAD). Its inclusion can affect the results of intensity-based image processing methods and needs to be identified and suppressed before further analysis. This paper proposes a novel relative dependency measure using the Rough Set Theory (RST) for the identification of the pectoral muscle in MLO mammograms. The pectoral muscle is identified using an automatic thresholding and connected component labeling algorithm. A dataset of 322 MLO mammograms from the MIAS database has been used for evaluation. Pectoral muscle detection results are evaluated in terms of the proportion of correctly assigned pixels.