Back to all news
Foundation Models Advance Ophthalmic Imaging AI: Review Highlights Milestones and Challenges
A systematic review evaluates five cutting-edge ophthalmic AI foundation models, highlighting their technical evolution and clinical trial evidence.
Key Details
- 1Five foundation models reviewed: RETFound, VisionFM, EyeCLIP, EyeFM, MIRAGE.
- 2Multimodal models now integrate up to 11 imaging modalities and clinical text (EyeCLIP).
- 3EyeFM is the only model with RCT-based clinical validation, improving diagnostic accuracy (92.2% vs. 75.4%) and reducing reporting time by 63.3 seconds.
- 4Other models validated only retrospectively; no multicenter or prospective validations except EyeFM.
- 5RETFound, VisionFM showed promising results in predicting systemic diseases from eye images, but results are associations, not causality.
- 6Current key limitations: lack of demographic diversity, limited prospective validation, and reliance on 2D data slices.
Why It Matters
This review signals a shift from technical model performance to clinical integration and real-world validation—critical steps for adoption in radiology and ophthalmology AI. The findings underscore the need for more robust, prospective, and diverse clinical studies before broad clinical translation.
Source
EurekAlert
Related News
•EurekAlert
AI Model Combines ECG and Blood Tests to Spot Heart Transplant Rejection
An AI model integrating ECG data and blood biomarkers predicts heart transplant rejection with high specificity, reducing the need for invasive biopsies.
•EurekAlert
NIH Grant Funds AI Tool for Lung Cancer Discovery at Weill Cornell
Dr. Olivier Elemento receives an NIH Pioneer Award to develop an AI-human hybrid platform for cancer research, initially focusing on lung cancer using imaging and tissue data.
•EurekAlert
Study Shows AI Vision Misses Human-Like Perceptual Illusions
A York University study finds that AI vision systems lack the human-like perceptual shifts seen in biological vision during visual illusions.