Breast Cancer Classification using Deep Convolutional Neural Network

Over the last decade, the demand for early diagnosis of breast cancer has resulted in new research avenues. According to the world health organization (WHO), a successful treatment plan can be provided to individuals suffering from breast cancer once the non-communicable disease is diagnosed at an early stage. An early diagnosis of cure disease can reduce mortality all over the world. Computer-Aided Diagnosis (CAD) tools are widely implemented to diagnose and detect different kinds of abnormalities. In the last few years, the use of the CAD system has become common to increase the accuracy in different research areas. The CAD systems have minimum human intervention and producing accurate results. In this study, we proposed a CAD technique for the diagnosis of breast cancer using a Deep Convolutional Neural Network followed by Softmax classifier. The proposed technique was tested on the Wisconsin Breast Cancer Datasets (WBCD). The proposed classifier produced an accuracy of 100% and 99.1% for two different datasets, which indicates effective diagnostic capabilities and promising results. Moreover, we test our proposed architecture with different train-test partitions.