An investigation of the breast cancer classification using various machine learning techniques

Abstract

It is an extremely cumbersome process to predict a disease based on the visual diagnosis of cell type with precision or accuracy, especially when multiple features are associated. Cancer is one such example where the phenomenon is very complex and also multiple features of cell types are involved. Breast cancer is a disease mostly affects female population and the number of affected people is highest among all cancer types in India. In the present investigation, various pattern recognition techniques were used for the classification of breast cancer using cell image processing. Under these pattern recognition techniques, cell image segmentation, texture based image feature extraction and subsequent classification of breast cancer cells was successfully performed. When four different machine learning techniques: Kth nearest neighbor (KNN), Artificial Neural Network ( ANN), Support Vector Machine (SVM) and Least Square Support Vector Machine (LS-SVM) was used to classify 81 cell images, it was observed from the results that the LS-SVM with both Radial Basis Function (RBF) and linear kernel classifiers demonstrated the highest classification rate of 95.3488% among four other classifiers while SVM with linear kernel resulted a classification rate of 93.02% which was close to LSSVM classifier. Thus, it was demonstrated that the LS-SVM classifier showed accuracy higher than other classifiers reported so far. Moreover, our classifier can classify the disease in a short period of time using only cell images unlike other approaches reported so far.