Intellectual detection and validation of automated mammogram breast cancer images by multi-class SVM using deep learning classification

Abstract

Breast cancer is an important cause of death in females. Early recognition of this disease with the assistance of mammography reduces the death rate. Deep learning (DL) is an approach being utilized and requested by radiologists to assist in making an accurate diagnosis, and it can help to improve outcome predictions. This paper includes a new approach, applied on the Mini-MIAS dataset of 322 images, involving a pre-processing method and inbuilt feature extraction using K-means clustering for Speed-Up Robust Features (SURF) selection. A new layer is added at the classification level, which carries out a ratio of 70% training to 30% testing of the deep neural network and Multiclass Support Vector Machine (MSVM). The outcome described herein demonstrates that the accuracy rate of the proposed automated DL method using K-means clustering with MSVM is improved as compared with a decision tree model. Experimental results show that the average accuracy (ACC) rates of the three classes, i.e., normal, benign and malignant cancer, using the proposed method, are 95%, 94% and 98%, respectively. The increased sensitivity rate is 3%, specificity is 2%, and Receiver Operating Characteristics (ROC) area is 0.99 using SVM compared to the Multi-Layer Perceptron (MLP) and J48+K-mean clustering WEKA manual approach. A 10-fold cross validation was used, and the obtained results for the Support Vector Machine (SVM), K-nearest neighbour (KNN), linear discriminant analysis (LDA) and Decision Tree were 96.9%, 93.8%, 89.7% and 88.7%, respectively.