Photographic age gap is associated with frailty and predicts survival in laboratory mice

Photographic “age gap” in laboratory mice is linked to frailty and predicts shorter survival, with a hazard ratio of 1.14 per 100 days for mice predicted older than their age.

• Why it matters: Aging studies need cheap, non-invasive, repeatable biomarkers, and it is unclear whether unexplained variation in photo-based age reflects underlying biology rather than just measurement noise.
• What they did: The study used 2,846 photographs of 1,292 mice across four strains and seven facilities, training a frozen DINOv2 model plus ridge regression to predict age, with testing on unseen facilities.
• The result: Absolute age transfer was poor (median absolute error 138 days), but relative ordering within sessions was strong; mice predicted older than their age showed higher frailty (+0.12 SD per 100 days) and shorter survival (hazard ratio 1.14 per 100 days).

AI summary of the abstract · may contain errors · not medical advice · read the paper.