Experimental investigation on breast tissue classification based on statistical feature extraction of mammograms.

An important approach for describing a region is to quantify its structure content. In this paper, the use of functions for computing texture based on statistical measures is described. Six textural features for mammogram images are defined. The segmentation based on these textures would classify the breast tissue under four categories. The algorithm evaluates the region properties of the mammogram image and thereby would classify the image under four important categories based on the intensity level of histograms. Experiments have been conducted on images of mini-MIAS database (Mammogram Image Analysis Society database (UK)). The breast tissue classification thus obtained is comparatively better than the other normal methods. The validation of the work has been done by visual inspection of the segmented image by an expert radiologist. This work is a part of developing a computer aided decision (CAD) system for early detection of breast cancer. The classification results agree with the standard specified by the ACR-BIRADS (American College of Radiology-Breat Imaging And Reporting Data Systems). The accuracy of classification has been found to be 80% as per the visual inspection by an expert radiologist.