Retinal photograph–based deep learning predicts biological age, and stratifies morbidity and mortality risk

AI Generated Summary*

A deep learning algorithm trained on 129,236 retinal images from 40,480 Korean adults learned to flag whether someone had reached age 65; the resulting score, RetiAGE, was tested in 56,301 UK Biobank participants over 10 years, during which 2,236 (4.0%) died. With PhenoAGE, a composite of age and blood markers, accounted for, the highest RetiAGE quartile carried hazard ratios of 1.67 for all-cause death, 2.42 for cardiovascular death and 1.60 for cancer death versus the lowest quartile, plus 1.39 and 1.18 for cardiovascular and cancer events. Alone the score gave a c-index of 0.70 for cardiovascular death, and adding it to chronological age or PhenoAGE models raised c-index by 1 to 2%. Stratification looked stronger in men, pending replication.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.