Improving breast cancer classification using histopathology images through deep learning

Breast cancer is a serious threat to human health. Image analysis of pathological sections of tumors has become an important means of detection. Assisting physicians in early detection of potential breast cancer has been a prominent topic in the field of medical image diagnostics. On this basis, a method based on depth neural network is proposed to solve the difficulties and limitations of manual extraction in traditional methods, and to automatically recognize breast disease images. On this basis, the improved methods of data enhancement and fine adjustment are adopted to solve the problem of small sample training based on depth neural network, improve the accuracy of image classification, and better meet the needs of the medical field.