Pre-Processing Algorithms on Digital X-ray Mammograms
Abdelali Elmoufidi · 2019
Breast cancer is one of the most typical sorts of cancer worldwide, and the most frequent cancer in women. Mammography remains the most effective tool for the early detection of breast cancer, as well as the systems of computer-aided diagnosis (CAD) is typically used as a second opinion by the radiologists. These CAD systems based on mammography images for automatic diagnosis breast cancer. Generally, the mammography images contain some noises and additional unnecessary objects, Such as: artifacts, identification labels, etc. These objects can disrupt the diagnostic process partially or completely. Therefore, a technique of preparing these images before using it is mandatory. So, in this article, we propose a method for automatic preprocessing image. Our method consists of five steps. Firstly suppressed the additional unnecessary objects, next by removing the additional black background, third find the orientation of mammograms, fourth segmented the pectoral muscle, and we end by contrast enhancement. The proposed method is tested on DDSM (Digital database for Screening Mammography) database, consisting of 10480 mammograms, and Mini-MIAS (Mammogram Image Analysis Society, UK) database, consisting of 322 mammograms. The obtained results confirm the efficacy of the proposed method.