A foundation model for generalizable disease detection from retinal images

Yukun Zhou (), Mark A. Chia, Siegfried K. Wagner, Murat S. Ayhan, Dominic J. Williamson, Robbert R. Struyven, Timing Liu, Moucheng Xu, Mateo G. Lozano, Peter Woodward-Court, Yuka Kihara, Andre Altmann, Aaron Y. Lee, Eric J. Topol, Alastair K. Denniston, Daniel C. Alexander and Pearse A. Keane ()
Additional contact information
Yukun Zhou: University College London
Mark A. Chia: NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust
Siegfried K. Wagner: NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust
Murat S. Ayhan: University College London
Dominic J. Williamson: University College London
Robbert R. Struyven: University College London
Timing Liu: NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust
Moucheng Xu: University College London
Mateo G. Lozano: NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust
Peter Woodward-Court: University College London
Yuka Kihara: University of Washington
Andre Altmann: University College London
Aaron Y. Lee: University of Washington
Eric J. Topol: Scripps Research
Alastair K. Denniston: University of Birmingham
Daniel C. Alexander: University College London
Pearse A. Keane: NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust

Nature, 2023, vol. 622, issue 7981, 156-163

Abstract: Abstract Medical artificial intelligence (AI) offers great potential for recognizing signs of health conditions in retinal images and expediting the diagnosis of eye diseases and systemic disorders1. However, the development of AI models requires substantial annotation and models are usually task-specific with limited generalizability to different clinical applications2. Here, we present RETFound, a foundation model for retinal images that learns generalizable representations from unlabelled retinal images and provides a basis for label-efficient model adaptation in several applications. Specifically, RETFound is trained on 1.6 million unlabelled retinal images by means of self-supervised learning and then adapted to disease detection tasks with explicit labels. We show that adapted RETFound consistently outperforms several comparison models in the diagnosis and prognosis of sight-threatening eye diseases, as well as incident prediction of complex systemic disorders such as heart failure and myocardial infarction with fewer labelled data. RETFound provides a generalizable solution to improve model performance and alleviate the annotation workload of experts to enable broad clinical AI applications from retinal imaging.

Date: 2023
References: Add references at CitEc
Citations: View citations in EconPapers (6)

Downloads: (external link)
https://www.nature.com/articles/s41586-023-06555-x Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:nature:v:622:y:2023:i:7981:d:10.1038_s41586-023-06555-x

Ordering information: This journal article can be ordered from
https://www.nature.com/

DOI: 10.1038/s41586-023-06555-x

Access Statistics for this article

Nature is currently edited by Magdalena Skipper

More articles in Nature from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().