Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Learn about groundbreaking advances in AI-powered ophthalmology through this conference presentation by Pearse Keane and Yukun Zhou from UCL. Discover how RETFound, the first foundation model in ophthalmology, was developed using self-supervised learning on 1.6 million retinal images with a vision transformer architecture. Explore the model's superior performance across diverse clinical tasks including diabetic retinopathy screening, age-related macular degeneration progression prediction, and oculomics applications that use the eye as a window to systemic disease. Understand how this Nature-published research demonstrates improved robustness on external validation, enhanced fairness across ethnicities, and greater label efficiency for rare retinal diseases. Gain insights into the development process of this foundation model and future plans to scale from 2 million to 20 million images while incorporating 3D and multi-modal capabilities. Learn from Professor Pearse Keane, a leading expert in artificial medical intelligence and consultant ophthalmologist who initiated the collaboration between Moorfields Eye Hospital and Google DeepMind, and PhD researcher Yukun Zhou, creator of the open-source AutoMorph tool and recipient of UCL's inaugural Open Science and Scholarship Award.
