跳到正文
原文
Nature 健康护理主题·· 2026-10-16AI 评分47

Stanford 提出通用文本引导模型 MediSyn:跨 6 个专科与 10 种模态生成医学图像

A generalist model for diverse text-guided medical image synthesis

AI 导读

Stanford 团队在 npj Digital Medicine 发表开放获取研究,提出基于公开数据训练的通用文本引导潜在扩散模型 MediSyn。该模型可跨 6 个医学专科和 10 种影像模态生成图像,在计算效率上优于特定任务模型套件。研究显示该模型能提升数据有限场景下的分类器性能,验证了通用生成模型在加速医学算法研发中的潜力。

正文

npj Digital Medicine (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

Deep learning requires extensive data to achieve robust performance. However, data availability is often restricted in medicine due to patient privacy concerns. Synthetic data from medical image generative models present a possible solution to these challenges. Unfortunately, many models are often task-specific or trained on private datasets, limiting scalability and reproducibility. To address this, we introduce MediSyn: an open-access, generalist, text-guided latent diffusion model. MediSyn can generate images across 6 medical specialties and 10 imaging modalities, while being trained exclusively on public data. Our study provides several key contributions using MediSyn. First, we demonstrate that a generalist model can generate images of comparable quality to task-specific models across multiple specialties. Second, we show that this generalist approach is more computationally efficient than a suite of task-specific models. Third, we establish that a generalist model can produce realistic, text-aligned images across visually and medically distinct modalities. Fourth, we empirically demonstrate that these synthetic images are visually distinct from real images used in training. Finally, we show that a generalist model can produce images that improve classifier performance in data-limited settings across multiple specialties. Altogether, our findings highlight the potential of generalist image generative models to accelerate medical algorithmic research and development.

Acknowledgements

We would like to thank the Stanford Sherlock cluster for providing computational resources and support that contributed to these research results. This work was supported, in part, by awards from the National Heart, Lung, and Blood Institute (1R01HL157235-01A1) and the American Heart Association (25IPA1454136) to W.H. The funders had no role in the design, data collection, analysis, or interpretation of this study.

Funding

This work was supported, in part, by awards from the National Heart, Lung, and Blood Institute (1R01HL157235-01A1) and the American Heart Association (25IPA1454136) to WH.

Author information

Author notes

  1. These authors contributed equally: Joseph Cho, Mrudang Mathur, Cyril Zakka.

Authors and Affiliations

  1. Department of Cardiothoracic Surgery, Stanford University, Stanford, CA, USA

    Joseph Cho, Mrudang Mathur, Cyril Zakka, Dhamanpreet Kaur, Matthew Leipzig, Kelly Higa, Alex Dalal, Aravind Krishnan, Matthew Duda, Ashley Choi, Robyn Fong & William Hiesinger

  2. Department of Ophthalmology, Stanford University, Stanford, CA, USA

    Eubee Koo, Karen Wai, Cindy S. Zhao, Ehsan Rahimy & Lyna Azzouz

  3. Department of Radiology, Stanford University, Stanford, CA, USA

    Akshay Chaudhari

  4. Department of Biomedical Data Science, Stanford University, Stanford, CA, USA

    Akshay Chaudhari

  5. Division of Cardiovascular Surgery, University of Pennsylvania, Philadelphia, PA, USA

    Rohan Shad

Authors

  1. Joseph Cho
  2. Mrudang Mathur
  3. Cyril Zakka
  4. Dhamanpreet Kaur
  5. Matthew Leipzig
  6. Kelly Higa
  7. Alex Dalal
  8. Aravind Krishnan
  9. Eubee Koo
  10. Karen Wai
  11. Cindy S. Zhao
  12. Akshay Chaudhari
  13. Matthew Duda
  14. Ashley Choi
  15. Ehsan Rahimy
  16. Lyna Azzouz
  17. Robyn Fong
  18. Rohan Shad
  19. William Hiesinger

Corresponding authors

Correspondence to Joseph Cho or William Hiesinger.

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/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cho, J., Mathur, M., Zakka, C. et al. A generalist model for diverse text-guided medical image synthesis. npj Digit. Med. (2026). https://doi.org/10.1038/s41746-026-03381-z

Download citation

  • Received: 20 April 2026

  • Accepted: 28 September 2026

  • Published: 16 October 2026

  • DOI: https://doi.org/10.1038/s41746-026-03381-z

来源:Nature 健康护理主题 · nature.com