SIFT based Feature Extraction in Breast Image using Soft Computing

The American Cancer Society (ACS) recommends women aged 40 and above to have a mammogram every year and calls it a gold standard for breast cancer detection. In this work a method for categorization of breast tissue thickness from mammographic images is proposed. The purpose of the method is to determine which class the breast tissue belongs to namely, fatty, fatty-glandular and dense-glandular. The SIFT algorithm is introduced for this purpose and it is used as the local feature extraction method, and k-NN algorithm is used for supervised classification. The SIFT features of each class can effectively model the breast tissue and the classification accuracy over 90% is achieved by classifier. Local adaptive threshold technique is used for the segmentation of cancer tissue. This examination aims at providing an outline about recent advances and developments in the field of breast cancer using mammograms, specifically focusing on the numerical aspects, aiming to act as a mathematical briefing for intermediates and experts in the field.