大型推理模型在临床诊断中的准确性-安全性权衡
The accuracy-safety trade-off of large reasoning models in clinical diagnostics
基于真实电子健康记录构建的AIMS-LDR基准显示,大型推理模型(LRM)较标准大语言模型(LLM)假阳性诊断增加逾34%、可重复性约低40%,但主要诊断更准确、鉴别诊断更具临床相关性。15名医师验证中,LRM扩展诊断覆盖、医师核验可靠性的结构化协作使诊断准确率提高20%、假阳性减少44%。作者称该结果来自受控回顾性条件,临床部署前仍需前瞻性验证。
- Article
- Open access
- Published: 29 October 2026
- Shuai Wu ORCID: orcid.org/0009-0001-7528-67821 na1,
- Linwei Chen2,3 na1,
- Yunxiang Fu1,
- Xiaoman Zhang ORCID: orcid.org/0000-0002-7696-93664,
- Luyang Luo ORCID: orcid.org/0000-0002-7485-41514,
- Meng Lou ORCID: orcid.org/0000-0001-5409-02811,
- Yangxi Huang5,
- Guanhua Chen6,
- Ying Fu7,
- Ya Xing Wang8,
- Qian Di9,
- Yuyin Zhou ORCID: orcid.org/0000-0003-2232-956310,
- Xiang Li ORCID: orcid.org/0000-0002-9851-637611,
- Tien Yin Wong ORCID: orcid.org/0000-0002-8448-12643,8,
- Yizhou Yu ORCID: orcid.org/0000-0002-0470-55481,2 &
- …
- Hong-Yu Zhou ORCID: orcid.org/0000-0002-1256-70503
Nature Communications (2026) Cite this article
We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.
Abstract
Large reasoning models (LRMs) offer potential advantages for clinical diagnosis over standard large language models (LLMs), yet their utility and safety remain unclear. Here we show, on AIMS-LDR, a long-context diagnostic reasoning benchmark built upon real-world electronic health records, covering six matched model pairs and 12 specialties, that LRMs generate over 34% more false positive diagnoses and show approximately 40% lower reproducibility than their LLM counterparts, while making more accurate primary diagnoses and more clinically relevant differentials. Experiments confirm this trade-off across prompt designs, model hyperparameters, and evaluation paradigms. A structured collaboration model, in which the LRM broadens diagnostic coverage and the clinician verifies reliability, increases diagnostic accuracy by 20% and reduces false positives by 44% in a 15-physician validation study. These results, obtained under controlled retrospective conditions, suggest that pairing LRM reasoning with physician oversight may support safer and more accurate diagnostic decision-making, warranting prospective validation before clinical deployment.
Acknowledgements
The authors thank the developers and maintainers of the MIMIC-IV and eICU Collaborative Research Database resources, as well as the PhysioNet platform, for making these de-identified clinical datasets available for research.
Funding
S.W., Yunxiang F., M.L. and Y.Y. were supported by the Hong Kong Research Grants Council through the General Research Fund (grant no. 17207722).
Author information
Author notes
These authors contributed equally: Shuai Wu, Linwei Chen.
Authors and Affiliations
School of Computing and Data Science, The University of Hong Kong, Hong Kong SAR, China
Shuai Wu, Yunxiang Fu, Meng Lou & Yizhou Yu
Hong Kong Generative AI Research & Development Center, Hong Kong SAR, China
Linwei Chen & Yizhou Yu
School of Biomedical Engineering, Tsinghua Medicine, Tsinghua University, Beijing, China
Linwei Chen, Tien Yin Wong & Hong-Yu Zhou
Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
Xiaoman Zhang & Luyang Luo
School of Nursing, LKS Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, China
Yangxi Huang
Department of Statistics and Data Science, Southern University of Science and Technology, Shenzhen, China
Guanhua Chen
School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China
Ying Fu
Beijing Visual Science and Translational Eye Research Institute (BERI), School of Clinical Medicine, Beijing Tsinghua Changgung Hospital Eye Center, Tsinghua Medicine, Tsinghua University, Beijing, China
Ya Xing Wang & Tien Yin Wong
Vanke School of Public Health, Tsinghua University, Beijing, China
Qian Di
Department of Computer Science and Engineering, University of California, Santa Cruz, CA, USA
Yuyin Zhou
Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA
Xiang Li
Authors
- Shuai Wu
- Linwei Chen
- Yunxiang Fu
- Xiaoman Zhang
- Luyang Luo
- Meng Lou
- Yangxi Huang
- Guanhua Chen
- Ying Fu
- Ya Xing Wang
- Qian Di
- Yuyin Zhou
- Xiang Li
- Tien Yin Wong
- Yizhou Yu
- Hong-Yu Zhou
Corresponding authors
Correspondence to Tien Yin Wong, Yizhou Yu or Hong-Yu Zhou.
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Wu, S., Chen, L., Fu, Y. et al. The accuracy-safety trade-off of large reasoning models in clinical diagnostics. Nat Commun (2026). https://doi.org/10.1038/s41467-026-78520-x
Received: 18 November 2025
Accepted: 28 September 2026
Published: 29 October 2026
DOI: https://doi.org/10.1038/s41467-026-78520-x
来源:Nature 健康护理主题 · nature.com