Breast Cancer Detection Using Deep Neural Network (DNN) on Histopathological Data

The leading cause of death for women in both developed and less developed nations is breast cancer (BC). It may be possible to treat cancer patients effectively by classifying the subtype of malignant (cancer) lesions. With the recent development in computer vision and deep learning, convolution neural networks (CNN) achieved enormous success in image classification and widely used in medical image processing. To recognize the subtype of cancer automatically of the whole slide images (WSI), which is computational impossible. The proposed Multi-Scale, Multi-Channel feature network for breast histopathological image classification follows the knowledge sharing strategy by sharing learned features at each stream across the network and the attention mechanism. The proposed module achieved accuracy for different magnification factors (MF) (40×, 100×, 200× and 400×) but the superior accuracy i.e. 99.25% for multi-class and 99.70% for binary at 200× MF. We observed that Deep Neural Network (DNN) model were better than existing models like VGG16, Xception and ResNet152, MuSCF-Net.