University of Lausanne AI estimates biological age via retinal photos

Researchers at the University of Lausanne have developed an artificial intelligence algorithm capable of estimating a person's biological age and assessing disease risks by analyzing retinal photographs. Published in the journal ‘Nature Communications’, the study utilized images from over 70,000 individuals aged 40 to 79. The AI model estimates retinal age with an average deviation of less than three years from chronological age.

A significant discrepancy between biological and actual age can indicate increased risks for cardiovascular and respiratory diseases, cancer, and dementia. The AI identifies subtle features nearly invisible to the human eye, such as minute variations in color, brightness, and blood vessel density.

The study also revealed gender-based differences in how retinal age correlates with health. In men, an older-looking retina was more strongly linked to metabolic syndrome markers, including high blood pressure, diabetes, and obesity. In women, retinal age was more closely associated with vascular factors, such as thrombosis risk. Additionally, the research noted that menopause appears to influence these patterns, as women's retinal age relative to men shifts after the onset of menopause.