EATNet: An extensive attention-based approach for cervical precancerous lesions diagnosis in histopathological images

Jing Xu, Lei Shi, Yameng Zhang, Guohua Zhao, Yufei Gao

Abstract: Highlights•Extensive ATtention Network to learn representations from whole slide images.•Extending the bag-of-words strategy and enabling to train in end-to-end mode.•Multi-scale dependencies encoding captures clinically relevant representations.•Bottom-up decoding and attention aggregation reduce diagnostic uncertainty.•EATNet achieves a reasonable trade-off between performance and model complexity.