Breast cancer is one of the most common cancers and it leads to death among women. Diagnosis of breast cancer
manually requires more time and highly experienced pathologists. For the improvement of diagnostic consistency and
efficiency, the computer assisted diagnosis systems are used in order to overcome this problem. The method proposed in this
paper includes image segmentation using sliding window mechanism, extracting feature vectors by using local binary pattern
algorithm and classification of histology images into benign and malignant classes using support vector machine. The
approach here is applied to perform classification of breast cancer histology images and achieve certain accuracy. In this work
the used algorithms are compared with various other methodologies where the proposed methodology obtain better accuracy
rate and In ROC analysis, the results also contain an optimal classification performance value as the area under curve (AUC)
of 0.90 which is more efficient compared to others.
