A-Eye

When light floods into our eyes, it’s the dome-like retina at the back, studded with light-sensitive photoreceptors, that collects the information. Our retinas need a healthy blood supply, and this is disturbed in conditions like macular degeneration. Here generative artificial intelligence helps to design one of hundreds of realistic but virtual retinas – networks of tiny artificial arteries (red) and veins (blue) used to train algorithms to segment (spot patterns) in the web of vessels. Later in the video, researchers use the synthetic retina to simulate the behaviour of fluorescein, a chemical commonly used to highlight issues with circulation around the retina. The fully trained AI eye can now spot patterns and problems in pictures of real retinas and may soon be put to work is hospitals and universities. Such methods could speed up early detection and diagnosis of retinal issues and therefore treatment for patients.

Written by John Ankers

John Ankers

Writer

A former researcher in systems biology, John is now an online biology tutor and professional coach supporting the wellbeing of scientists and NHS professionals. He's writing, too, of course.

Website

• Video from work by Emmeline E. Brown and colleagues
• Centre for Computational Medicine, University College London, London, UK
• Video originally published with a Creative Commons Attribution 4.0 International (CC BY 4.0)
• Published in Nature Communications, August 2024