A team from the Department of Computational Biology (Faculty of Biology and Medicine) at UNIL, led by Prof. Sven Bergmann, has developed an AI model that estimates a person’s “retinal age” from a simple photograph of the back of the eye. Trained on images of more than 70,000 people aged 40 to 79 from an English biobank, it is off by less than three years on average.
The gap between retinal age and actual age is associated with many aspects of health, including cardiovascular and respiratory diseases, cancer and dementia, as well as with health events occurring over the following fifteen years. The study, whose first author is Olga Trofimova, was carried out with the Jules-Gonin Eye Hospital and the Erasmus University Medical Center in Rotterdam. It also reveals differences between the sexes: before menopause, women have retinas that appear younger than those of men, and this reverses afterwards. Smoking is the factor most strongly associated with a higher retinal age. The results are published in Nature Communications.