A groundbreaking study led by researchers from the Yong Loo Lin School of Medicine at the National University of Singapore (NUS Medicine) has shown that artificial intelligence can analyze standard retinal photographs to predict an individual’s risk of cognitive decline and dementia. Published in Alzheimer's & Dementia, the research introduces a new deep-learning biomarker called RetiPhenoAge, which estimates the biological age of the retina based on routine eye scans. The study involved over 500 participants from memory clinics in Singapore and found that those with a higher retinal biological age, as measured by RetiPhenoAge, faced a 25% to 40% increased risk of developing cognitive decline or dementia within five years for each standard deviation increase in the biomarker. These results were validated in a much larger cohort of over 33,000 individuals from the UK Biobank, where elevated RetiPhenoAge also predicted a higher dementia risk over a 12-year follow-up period, confirming its reliability across different populations. The researchers linked RetiPhenoAge to key biological processes tied to neurodegeneration. Brain imaging and blood tests showed that higher retinal aging scores corresponded with brain changes and alterations in blood proteins associated with aging and cognitive decline, suggesting that the retina may serve as a window into brain health. Professor Cheng Ching-Yu, director of the Center for Innovation and Precision Eye Health at NUS Medicine, emphasized the potential of RetiPhenoAge for early detection. He noted that the tool enables non-invasive estimation of biological age, helping clinicians identify at-risk individuals before symptoms emerge, which could lead to earlier interventions and better outcomes. Professor Christopher Chen, deputy chair of the Healthy Longevity Translational Research Programme at NUS Medicine and director of the Memory, Aging and Cognition Centre at NUHS, highlighted the scalability and affordability of the approach. He said RetiPhenoAge could become a key component of community-wide dementia screening, especially as dementia cases rise globally. The technology uses existing retinal imaging equipment found in many polyclinics, making it easy to integrate into routine health check-ups. Dr. Sim Ming Ann and Asst Prof Tham Yih Chung, co-first authors of the study, expressed hope that the findings would lead to improved care by enabling earlier identification of at-risk individuals. Dr. Sim is a consultant at Ng Teng Fong General Hospital and NUH and a Ph.D. candidate at NUS Medicine, while Asst Prof Tham is affiliated with the Center for Innovation and Precision Eye Health. The research team is now expanding validation efforts across diverse populations in Asia and beyond. They are also exploring how RetiPhenoAge can be used to monitor responses to interventions such as lifestyle changes, medications, and other therapies aimed at slowing or preventing cognitive decline. This work could pave the way for a new era of accessible, early-stage dementia screening and personalized brain health management.