Mammographic mass classification using Gabor Wavelet based features of circular scan lines

Breast cancer develops from breast tissue. This cancer is reported as the second most deadly cancer in the world and the most common cancer in most cities as well as in rural areas of India. Early detection can play an effective role in prevention and cure. At present the most reliable detection technology is digital mammography. At the early stages of breast cancer, it is very difficult to detect as the clinical signs are very mild and vary in appearance. Therefore, automatic detection by medical images by computer-Aided Detection (CAD) systems becomes highly desirable. This paper aims to design and develop computer-Aided Diagnosis (CADx) system for mass classification in digital mammograms. A comparative analysis is performed between wavelet based features and Gabor features in application to mammographic mass classification. The Gabor feature is found to have the ability to classify the benign and malignant masses more accurately than the wavelet based circular scan line features proposed earlier.