Identification of suspicious masses in mammography

A study on mammographic images is proposed to characterize non suspicious zones and cancerous ones. In particular, the attention has been devoted to ductal carcinoma. It is one of the leading cause of death for women and early diagnosis significantly reduces breast cancer mortality. In this paper properties like mean values, standard deviation, correlation, contrast, energy and homogeneity are evaluated. The images are segmented on the basis of this information in order to identify and enhance suspicious zones. These zones are characterized also by their morphological properties, such as their extension and the raggedness of their contours. The proposed analysis has been tested on mammographic images for which the diagnosis of an expert was available. The numerical results show a significant success in cancerous zones identification.