AI-Driven Eye Scans Predict Risk of Cognitive Decline and Dementia

A groundbreaking study conducted by researchers from the Yong Loo Lin School of Medicine at the National University of Singapore (NUS Medicine) has demonstrated that artificial intelligence (AI) analysis of retinal images can effectively forecast an individual's likelihood of experiencing cognitive decline and developing dementia. Published in the journal Alzheimer's & Dementia, this innovative research marks the first in Singapore to utilize retinal aging biomarkers as a potential tool for early brain health assessment.

The team, led by Professor Cheng Ching-Yu of the Center for Innovation and Precision Eye Health and Professor Christopher Chen of the Healthy Longevity Translational Research Programme, developed a deep-learning biomarker named RetiPhenoAge. This biomarker estimates the biological age of the retina by analyzing standard eye photographs, providing insights into the aging process of neural tissues linked to cognitive health.

The study involved over 500 participants from memory clinics in Singapore. Results revealed that individuals with a higher retinal biological age had a significantly increased risk—ranging from 25% to 40% per standard deviation increase in RetiPhenoAge—of experiencing cognitive decline or dementia within five years. These findings were further validated through data from more than 33,000 participants in the UK Biobank, where elevated RetiPhenoAge was similarly associated with a heightened risk of developing dementia over a 12-year follow-up period. The associations between retinal aging, brain changes, and blood biomarkers support the potential of using retinal scans as a non-invasive proxy for neurodegenerative processes.

Prof. Cheng highlighted that RetiPhenoAge allows healthcare providers to estimate biological age without invasive procedures, enabling earlier identification of those at risk. This could lead to targeted interventions before clinical symptoms manifest. The research team aims to expand validation efforts across diverse populations in Asia and explore clinical applications for early diagnosis and risk stratification.

As dementia cases continue to rise globally, the development of scalable and affordable screening tools like RetiPhenoAge becomes crucial. Researchers are also investigating how retinal imaging could monitor individual responses to preventative strategies such as lifestyle changes or lifestyle modifications and pharmacological treatments. Future plans include community-based screening initiatives and tracking therapeutic responses in at-risk populations.

This pioneering advancement in digital biomarkers demonstrates the synergy of AI and non-invasive imaging, promising a new frontier in early detection and management of cognitive health risks. RetiPhenoAge, utilizing existing retinal imaging technology available in many clinics, offers a practical and scalable solution for widespread screening and early intervention.

Source: Medical Xpress