Breast TransFG Plus: Transformer-based fine-grained classification model for breast cancer grading in Hematoxylin-Eosin stained pathological images

Zhencun Jiang, Zhengxin Dong, Jinfu Fan, Yang Yu, Yuanqing Xian, Zhongjie Wang

Abstract: Highlights•The proposed model is an transformer-based fine-grained classification model for breast cancer grading.•The feature distribution characteristic of pathological images is fully considered.•The proposed part selection module plus and double head classification structure can capture global features and key local features.•The accuracy of breast cancer grading is 99.39%, which is superior to previous studies.