Automated Detection of Breast Cancer Metastases

For women, breast cancer is the most commonly diagnosed cancer, which brings a heavy workload to pathologists. Because this diagnostic procedure is now prone to being time-consuming and sometimes misinterpreting. In order to solve these problems, techniques related to deep learning and machine learning have been applied to the diagnostic process of breast cancer. However, some problems have been found in application of these technologies, such as imbalanced data sets. This paper proposes a data augmentation technique based on generative adversarial networks (GAN) which can solve the problem of data imbalance, then uses ResNet to evaluate the impact of different data augmentation techniques on the experimental results.