Revolutionary AI Eye Scan Predicts Dementia Risk

A groundbreaking study conducted by researchers at the National University of Singapore has revealed that artificial intelligence (AI) can analyze retinal images to predict an individual's risk of cognitive decline and dementia. This pioneering research, published in the journal Alzheimer's & Dementia, establishes retinal aging biomarkers as a promising tool for managing brain health.

The research team, led by professors Cheng Ching-Yu and Christopher Chen, developed a deep learning biomarker known as RetiPhenoAge. This innovative tool assesses the biological age of the retina using standard eye photographs. The study involved over 500 participants from memory clinics in Singapore, revealing that individuals with a higher retinal biological age exhibited a significantly increased risk--between 25% and 40% per standard deviation increase in RetiPhenoAge--of developing cognitive decline or dementia within a five-year timeframe.

To further validate their findings, the team analyzed data from a larger cohort of more than 33,000 participants from the UK Biobank. The results indicated that a higher RetiPhenoAge was consistently associated with an elevated risk of dementia over a twelve-year follow-up period, underscoring its predictive potential across diverse populations.

The study also demonstrated that retinal aging reflects critical biological processes related to neurodegeneration. The researchers utilized brain scans and blood markers to confirm RetiPhenoAge's correlation with brain changes and age-related variations in blood proteins. This connection suggests retinal aging could serve as a viable proxy for assessing cognitive health.

Professor Cheng noted that RetiPhenoAge enables non-invasive estimation of an individual's biological age, providing essential insights for cognitive health management and broader aging research. This advancement may help healthcare providers identify individuals at risk of cognitive decline before symptoms manifest, allowing for targeted interventions.

Professor Chen emphasized the urgent need for scalable and predictive tools, especially with the global rise in dementia cases. RetiPhenoAge has the potential to be integrated into routine health screenings, making early detection of dementia risk more accessible and affordable.

The research team is committed to further validating this screening tool in larger and more diverse populations and assessing its effectiveness in clinical settings. Their ongoing efforts aim to enhance early treatment options for dementia.

Co-first authors of the study, Dr. Sim Ming Ann and Assistant Professor Tham Yih Chung, expressed hope that their findings will lead to improved care strategies, enabling doctors to identify individuals at risk of dementia sooner, thus facilitating earlier interventions and better patient outcomes.

This significant advancement in digital biomarkers illustrates the potential of combining AI with non-invasive imaging techniques to address pressing healthcare challenges. Given that RetiPhenoAge utilizes retinal scans from existing imaging technologies commonly available in Singapore's polyclinics, it presents a convenient and scalable solution for routine health checks. The researchers are now focused on validating the biomarker across various populations in Asia and beyond, while also exploring its applicability in community health settings.

Future research will involve using retinal imaging to screen for cognitive impairment within communities. The team is also investigating how RetiPhenoAge can track individual responses to interventions aimed at mitigating or preventing cognitive decline and dementia, including lifestyle changes, pharmacological treatments, and other therapeutic methods.