A deep learning approach for mitosis detection: Application in tumor proliferation prediction from whole slide images

作者：

Highlights：

• Fully automated end-to-end system for cancer grading from whole slide images via mitosis counting.

• Robust strategy to remove hard and very hard non-mitosis objects within the histopathology images.

• A deep pre-processing technique to increase the intensity difference between mitosis cell pixels and background.

• Reliable tool for tumor proliferation assessment.

摘要

•Fully automated end-to-end system for cancer grading from whole slide images via mitosis counting.•Robust strategy to remove hard and very hard non-mitosis objects within the histopathology images.•A deep pre-processing technique to increase the intensity difference between mitosis cell pixels and background.•Reliable tool for tumor proliferation assessment.

论文关键词：Breast cancer grading,Mitosis detection,Whole-slide image,Deep learning

论文评审过程：Received 13 May 2020, Revised 25 February 2021, Accepted 28 February 2021, Available online 6 March 2021, Version of Record 10 March 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102048