RETFound
Benefits
RETFound is an AI foundation model in ophthalmology that can identify sight-threatening eye diseases and predict other health conditions including heart attacks, stroke, and Parkinson's disease. It outperforms existing AI systems across complex clinical tasks, works well in diverse populations and patients with rare diseases, and requires only 10% of human labels compared to other systems. The model improves diagnosis of eye diseases like diabetic retinopathy and glaucoma, while also predicting systemic diseases.
Details
- Tool name
- RETFound
- Organisation
- Moorfields Eye Hospital NHS Foundation Trust
- Status
- live
- AI method (as recorded)
- Computer Vision, generativeAI, Machine Learning
- AI method (normalised tags)
- Computer Vision generativeAI Machine Learning
- Usecase
- Health AI
- Origin
- moorfields.nhs.uk
- Owning team
- Moorfields Eye Hospital and UCL Institute of Ophthalmology
- Date added
- 13 September 2023
- Scrape date
- Source note
- RETFound was developed using millions of eye scans from the NHS and is being made open-source and freely available on GitHub for any institution worldwide. It was trained on 1.6 million images from Moorfields Eye Hospital using AI tools provided by INSIGHT. The research was published in Nature and led by Yukun Zhou and Professor Pearse Keane. [Source]