Digital mammography texture analysis by computer assisted image processing

We investigate the use of mammograms texture features in a clinical evaluation in mammography in order to assist in the classification of an image based on the breast tissue index. Breast tissue indices give the possibility to analyze images and determine the similarity of the images and evaluate the classification of the obtained image The measures of similarity of images used for the texture analysis are based upon textural characteristics and the distribution of density in the breast. The features used include the gray-level histogram, and texture features based upon the gray-level co-occurrence matrix, moment-based features obtained by Matlab program application. The performance of the method was measured in terms of precision of retrieval measurements. The results indicate improvement that is related to breast density classification in mammograms. Image processing assisted by computer give the characterization of masses and micro-calcification texture. The methodology proposed in the present work is generic, and can be applied also to other imaging modalities and clinical applications, as for example, Ultrasound breast evaluation and Scintimammography.