Clinically informed intermediate reasoning enables generalizable prostate cancer prognostication through machine learning in limited settings
Machine learning has shown promise in medical image classification. However, its generalizability remains challenging. Here, we show that data-efficient pre-surgical prognostication of prostate cancer from biopsy specimens is enabled by versatile feature extraction from whole-mount histopathology and a clinically informed intermediate reasoning step. With data from multiple institutions, our pipeline resolved dual-domain shifts across specimen types and institutions and achieved consistent external validation, reinforced by comprehensive analyses of generalizability. This highlights the robustness of our prognostic approach when compared to the Gleason grading system. We establish an equitable, interpretable, and clinically applicable framework, supporting actionable decisions for prognosis and treatment planning, even in limited real-world clinical environments.
Authors
- Yoichiro Yamamoto (ORCID: https://orcid.org/0000-0003-0799-4338)
- Takuma Kato (ORCID: https://orcid.org/0000-0003-2047-496X)
- Hiromu Morikawa
- Kotaro Tsutsumi (ORCID: https://orcid.org/0000-0002-6576-0970)
- Akira Shimizu (ORCID: https://orcid.org/0000-0002-4364-9251)
- Yuka Toyama
- Toyonori Tsuzuki (ORCID: https://orcid.org/0000-0002-4855-4366)
- Jun Akatsuka (ORCID: https://orcid.org/0000-0002-2671-7064)
- Yukihiro Kondo (ORCID: https://orcid.org/0000-0003-4141-9230)
- Go Kimura (ORCID: https://orcid.org/0000-0003-0088-9324)
- Yuki Endo (ORCID: https://orcid.org/0009-0002-0787-9321)
- Hisamitsu Ide (ORCID: https://orcid.org/0000-0001-6350-9814)
- Naonori Ueda (ORCID: https://orcid.org/0000-0001-5701-9333)
- Atsushi Marugame
- Masao Ueki (ORCID: https://orcid.org/0000-0003-0468-7041)
- Ryuji Ohashi (ORCID: https://orcid.org/0000-0001-7053-038X)
- Naoto Sassa (ORCID: https://orcid.org/0000-0002-2424-6529)
- Junya Iwazaki
- Tomoharu Kiyuna (ORCID: https://orcid.org/0000-0003-3050-6718)
- Shigeo Horie (ORCID: https://orcid.org/0000-0002-8612-8368)
- Yasushi Numata
- Maki Ogura (ORCID: https://orcid.org/0000-0002-0380-9396)
- Toshiyuki China (ORCID: https://orcid.org/0009-0004-3217-9911)
- Takayuki Takahashi (ORCID: https://orcid.org/0000-0002-3457-0031)
- Kaori Ono
- Hayato Takeda (ORCID: https://orcid.org/0000-0002-4556-6601)
- Mami Takadate
Institutions
- NEC (Japan) (JP)
- Tohoku University (JP)
- Kagawa University (JP)
- University of California, Irvine (US)
- Juntendo University (JP)
- Aichi Medical University Hospital (JP)
- RIKEN Center for Advanced Intelligence Project (JP)
- Nagasaki University (JP)
- Nippon Medical School (JP)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2025-12-03
- DOI
- https://doi.org/10.1038/s41746-025-02193-x
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Japan Society for the Promotion of Science
- Japan Science and Technology Agency
- RIKEN
- Moonshot Research and Development Program