GLCM AND GLRLM BASED TEXTURE FEATURES FOR COMPUTER-AIDED BREAST CANCER DIAGNOSIS

This paper presents 15 texture features based on GL CM (Gray-Level Co-occurrence Matrix) and GLRLM (Gray-Level Run-Length Matrix) to be used in an automatic computer system for breast cancer diagnosis. The task of the system is to distinguish benign from malignant tumors based on analysis of fine needle biopsy micr oscopic images. The features were tested whether they provide impor tant diagnostic information. For this purpose the a uthors used a set of 550 real case medical images obtained from 50 pa tients of the Regional Hospital in Zielona Gora. Th e nuclei were isolated from other objects in the images using a h ybrid segmentation method based on adaptive thresholding and kmeans clustering. Described texture features were t hen extracted and used in the classification proced ure. Classification was performed using KNN classifier. Obtained results reaching 90% show that presented features are imp ortant and may significantly improve computer-aided breast cancer detection based on FNB images.