Particle.news
Download on the App Store

AI Analysis of Retinal Photos Predicts Osteoporosis Risk

Researchers report a low-cost eye-image signal that could support opportunistic screening.

Overview

  • The peer-reviewed study, published Thursday in PLOS Digital Health, links faster retinal aging to lower bone density and higher odds of later osteoporosis.
  • The team built a deep-learning model called RetiAGE using about 130,000 retinal images from a South Korean health study to estimate retinal biological age.
  • In 1,965 participants from Singapore’s PIONEER study, higher RetiAGE scores aligned with lower bone density on DEXA scans and higher fracture risk scores.
  • Validation in 43,938 UK Biobank participants found about 12% higher osteoporosis risk per standard deviation increase in RetiAGE, rising to about 25% among men.
  • Authors say retinal photography could offer a simple, noninvasive screening aid, though mechanisms, clinical thresholds, and real-world pathways still need testing.