Metastatic Breast Cancer Recognition in Histopathology Images Using Convolutional Neural Network with Attention Mechanism
Liang Yu, Jinglong Yang, Xiongwen Quan, Han Zhang · 2019
Lymph node tissue pathological analysis is one of the common methods for doctors to evaluate the types and stage of breast cancer. Using deep learning methods to detect breast cancer metastases has great research value. We proposes a model to automatically classify breast cancer metastases, using a convolutional neural network with attention mechanism. A Convolutional Block Attention Module is used in our network. We validate our model on PCam dataset, and obtain an AUC score of 0.976. These results demonstrate that using computer vision method has a great performance in pathological diagnoses.