Early Stage Breast Cancer Detection System using GLCM feature extraction and K-Nearest Neighbor (k-NN) on Mammography image

Abstract

The disease that threaten the women’s life is Breast Cancer. In medical imaging, the easier way to prevention the death caused by breast cancer is the early stage detection through computer aided diagnosis system. The aim of this study is to develop an early stage breast cancer detection system which can automatically classify abnormalities in mammograms. In this method, pre-processing stage is to remove the noise by using some median filter and then cropping is done on the image. Then, Otsu’s thresholding is used for segmentation breast region which are located on a non-uniform background. After performing the thresholding of the images, feature extraction is focused on the first order statistical and Gray Level Co-occurrence Matrix (GLCM) based textural features extraction techniques. Finally, all the extracted objects are classified whether they are normal or abnormal by using k-Nearest Neighbor (k-NN) classifier. The algorithm is validated by the mini - MIAS’s data base.