Pre-processing techniques to facilitate better detection of breast abnormalities using Digital Mammogram.

Digital Mammography is a popular technique for early, non-invasive, breast cancer detection. It is based on the identification and analysis of abnormalities in the breast tissue. Digital Mammography offers the potential for the development of algorithms that can enable computer-aided detection (CAD). This paper proposes such an algorithm based on a three-step process. The first step involves contrast enhancement using the Contrast Limited Adaptive Histogram Equalization (CLAHE) method. 2^nd Step is Isolation of Pectoral Muscle from ROI: The medial oblique view has been improved, and the breast and lymph nodes are now more visibly visible in it. The pectoral muscle is kept separate from the area of interest for further investigation (ROI). "Region growth" enables the process of grouping pixels or smaller portions in the image to merge and become larger sections or regions on the basis of some preconceived specifications. The fundamental method is to start from the tip known as the "seed", and then grow in the area by adding adjacent pixels. [ABSTRACT FROM AUTHOR]