An adaptive approach for computer aided screening of mammograms and classification of abnormalities

This paper aims for the development of a highly efficient computer aided decision system to automatically detect abnormalities in mammograms. Enhancement of the contrast of the intensity image by transforming the values using Contrast Limited Adaptive Histogram Equalization (CLAHE) is done for preprocessing of images after classifying the mammograms into various intensity levels. Then mathematical morphology is used for the extraction of abnormalities which are located on a non uniform background. After performing the thresholding of the image by extended maxima transformation by using adaptive H-domes transformation feature extraction is performed. Transformation constant (h) is based on the breast density of the mammogram considered. The Feature extraction is focused on the extraction of GLCM based statistical features of the objects. Finally the extracted objects are classified using Naive Baye's Classifier and abnormalities are detected. SVM classifier is also employed to classify the mammogram whether it is suspicious or not.