Seoul National University researchers develop AI retinal aging assessment

Researchers at Seoul National University in South Korea have developed a non-invasive method to assess biological aging using retinal images. The technology, known as Retinal Age Gap (RAG), utilizes artificial intelligence to estimate a person's age from the back of the eye and compares it to their actual chronological age.

The AI model was trained on 29,530 retinal images from 7,535 individuals and subsequently tested on 14,832 images from 7,416 different people. The system demonstrated high accuracy, predicting chronological age with an average error margin of only 2.5 to 2.7 years.

The study found correlations between a higher retinal age gap and various health factors. For instance, diabetes was associated with a 2.52-year increase in the gap, while smoking was linked to increases of approximately 0.50 years for active smokers and 0.46 years for former smokers. Other associations included macular degeneration and cataracts, though researchers noted these correlations do not necessarily imply direct cause-and-effect relationships.