Applicability of mitotic figure counting by deep learning: a development and pan‐cancer validation study
Mitotic figure counting is an established measure of cell proliferation that is included in grading systems. We developed a deep learning method for mitotic figure counting and evaluated its prognostic impact in multiple external validation datasets. The deep learning method was trained in whole slide images of tissue sections stained with haematoxylin and eosin from a publicly available breast cancer dataset where mitotic figures have been annotated by expert pathologists. The final model was externally validated according to a protocol with predefined analyses of 14 571 patient samples from 13 patient cohorts from seven different cancer types. The predefined primary analysis was univariable Cox survival analysis of the number of mitotic figures detected per mm<sup>2</sup>. Automatic mitotic figure counting correlated well with known proliferation rates, and patients with more mitotic figures per mm<sup>2</sup> had significantly worse patient outcome in all the studied cancer types except colorectal cancer. This study demonstrates the practical potential of automated, deep learning-based mitotic figure counting, both by automating pathology work and by suggesting expanded use in more cancer types, such as prostate cancer.
Authors
- Turid Torheim (ORCID: https://orcid.org/0000-0001-6191-2036)
- Tarjei S. Hveem
- Sepp de Raedt (ORCID: https://orcid.org/0000-0001-9291-1730)
- Gunnar B. Kristensen (ORCID: https://orcid.org/0000-0001-6231-7247)
- Bjørn Brennhovd
- Elin Richardsen (ORCID: https://orcid.org/0000-0003-2896-7910)
- Erik Skaaheim Haug (ORCID: https://orcid.org/0000-0003-2029-1901)
- Sigve Andersen (ORCID: https://orcid.org/0000-0002-2402-570X)
- Neil A. Shepherd (ORCID: https://orcid.org/0000-0003-4228-6580)
- Emma Rewcastle (ORCID: https://orcid.org/0000-0002-8943-3739)
- David Kerr (ORCID: https://orcid.org/0000-0001-7811-6722)
- Hanne A. Askautrud (ORCID: https://orcid.org/0000-0002-4404-3622)
- Vebjørn Kvikstad (ORCID: https://orcid.org/0000-0003-2980-8217)
- Frida Marie Ihle Julbø
- Knut Liestøl (ORCID: https://orcid.org/0000-0002-7929-582X)
- Melinda Lillesand (ORCID: https://orcid.org/0000-0001-8898-1330)
- Karolina Cyll (ORCID: https://orcid.org/0000-0001-6743-0724)
- Wanja Kildal
- John Arne Nesheim
- Maria Xepapadakis Isaksen
- J. Kalsnes
- R. Scott Jones (ORCID: https://orcid.org/0000-0003-3098-6248)
- T. Dønnem
- Hans Martin Mohn
- H E Danielsen
- L. T. Busund
- Andreas Kleppe
- Kristina Lindemann
- Manohar Pradhan
- Marco Novelli
- Emiel Janssen
- Odd Terje Brustugun
- Ole‐Johan Skrede
Institutions
- Oslo University Hospital (NO)
- University College Hospital (GB)
- University Hospital of North Norway (NO)
- Cheltenham General Hospital (GB)
- University of Liverpool (GB)
- University of Oslo (NO)
- Aintree University Hospitals NHS Foundation Trust (GB)
- Stavanger University Hospital (NO)
- Vestre Viken Hospital Trust (NO)
- Nuffield Orthopaedic Centre (GB)
- Sykehuset i Vestfold (NO)
- University of Oxford (GB)
- Centre for Arctic Gas Hydrate, Environment and Climate (NO)
- University College London (GB)
- UiT The Arctic University of Norway (NO)
- University of Stavanger (NO)
Publication Details
- Journal
- FEBS Open Bio
- Published
- 2026-02-12
- DOI
- https://doi.org/10.1002/2211-5463.70210
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Norges Forskningsråd