Abstract:
Breast cancer is one of the most common malignancies in women and a rare malignancy in men. It has been widely reported that breast cancer has become the second leading cause of cancer death among women. Over a lifetime, one in nine women risk contracting breast cancer. However, women who are diagnosed at an early stage can survive this often deadly disease. Mammography provides the best screening modality for detecting early breast cancer, even before a lesion is palpable. Because of the malignant mass pathology, the shape of the mammographic mass can be used to discriminate between malignant and benign masses. In this study geometric parameters such as area, perimeter, circularity, normalized circularity, radial distance mean and standard deviation, area ratio, orientation, eccentricity, moment invariants and Fourier descriptors up to ten, are calculated. The process starts with a segmentation phase, in which an expert radiologist segments the mammographic mass shapes within the mammographic database set. These pre-segmented mammographic mass shapes are then processed by a mass boundary detection algorithm to obtain descriptive geometric parameters. A carefully designed classification scheme is used in the final step to classify masses as benign or malign. The results show that normalized circulatory area and the Fourier descriptors can be used successfully for feature extraction. The software developed utilizes this finding in the automatic classification of the suspicious masses. mammogram database designed to store the images of the masses, calculated shape descriptor parameters and some additional data, such as patient history, category of the mass and biopsy report, if performed, which are required in BI-RADS is also introduced. The developed database is designed to be an Open Database Connectivity compliant relational database to support some future uses, such as screening the growth of suspicious masses, telemedical service support for sharing mass information and for facilitating statistical data analysis. A touch on memory system has been used as a tool to permit secure access to the electronic patient record in the mammogram database. The software is written in Delphi and runs on machines equipped with MS Windows.
