[REVISION HISTORY]

Seoul National University cognitive health research

Updated 2 times since CLSTR started tracking revisions of this situation.

What changed

2026-09-24 10:42 UTC → 2026-09-27 14:25 UTC · added ~~removed~~

Researchers at Seoul National University have developed new methods to identify cognitive decline and dementia risk through non-invasive assessments. One study focused on road safety by identifying a ‘silent risk group’ of elderly drivers. By analyzing cognitive performance on a ‘symbol connection test’ alongside education levels, researchers found they could identify high-risk drivers who may not realize their own declining abilities. The study noted these metrics are intended for precise ability assessments rather than restricting driving rights. Advancements in retinal imaging have further expanded these diagnostic capabilities. A research team led by Seoul National University Hospital developed an artificial intelligence model, published in ‘npj Digital Medicine’, that screens for current dementia and predicts future risk using retinal fundus photographs. Utilizing the ‘RETFound-MAE’ foundation model and ‘explainable AI’ technology, the system uses heatmaps to visualize retinal areas, such as the optic disc, that influence its decisions. This model demonstrated an AUROC of 0.750 for current dementia screening and a C-index of 0.812 for predicting future risk. Additionally, researchers developed a method to assess biological aging through the ‘Retinal Age Gap’ (RAG). This AI-driven technology estimates a person’s biological age from retinal images and compares it to their chronological age, achieving an average error margin of 2.5 to 2.7 years. The study identified correlations between a higher retinal age gap and factors such as diabetes, smoking, macular degeneration, and cataracts. Furthering this research, an AI system trained on approximately 30,000 retina scans has been shown to predict chronological age with high precision. Experts suggest that because the retina’s vascular structure is linked to the brain and cardiovascular system, this technology could be integrated into routine eye examinations to serve as a practical indicator for the early diagnosis of cardiovascular risks and systemic internal diseases before clinical symptoms appear.

Versions

Only revisions since CLSTR began indexing content versions appear here. Select a version to see what changed compared to the one before it.