Nature Communications · 26 August 2026 · From issue 27
Deep learning aging marker from retinal images unveils sex-specific clinical and genetic signatures
An eye scan predicted age within 2.85 years and flagged disease risk differently in men and women.
Olga Trofimova, Leah Böttger et al.
A computer program can look at a photo of the back of your eye and guess your age with a small margin of error, off by less than three years on average. When the guessed age is older than the real age, that gap links to higher risks of heart disease, dementia, cancer, and early death. This eye-age gap shows up differently in men and women. In men it connects more to metabolic problems, while in women it connects more to changes in eye blood vessels, especially after menopause.
The average error when the program guessed a person's age from an eye photo.
One large biobank study only, not tested elsewhere yet.
Integrity screen: passed (3 checks) Checked 27 August 2026. Retraction record: none. DOI resolves at doi.org. Metadata record found (Nature Communications). Read the source
This screen checks process signals. It does not tell you the findings are true.