A UK Biobank study utilising Artificial Intelligence has moved the needle closer to developing a retinal scan that can assess your stroke risk and possibly other health issues.

If eventually approved, it would mean that a simple, non-invasive eye scan could provide a significant indicator for a medical event that affects over 100 million people worldwide each year. Nearly 90 percent of occurrences are attributable to modifiable risk factors like hypertension, high cholesterol, diet and smoking.

Previous studies have shown that retinal microvascular abnormalities in the form of tortuosity, venous calibre, arteriovenous nicking and microaneurysms reflect damage from systemic conditions such as hypertension, diabetes and hypercholesterolaemia, all of which are known modifiable risk factors for stroke.

Researchers have identified a significant link between the structure of retinal blood vessels—the tiny vessels at the back of the eye—and the likelihood of future strokes. The findings not only offer a potential method for early detection but also highlight the critical role of eye health in revealing broader systemic conditions.

By analysing the patterns and changes in these vessels, scientists are paving the way for new preventive strategies, aiming to reduce the impact of strokes and save lives through earlier intervention.

Using data from the UK Biobank, a large-scale health database, researchers analysed retinal images of over 45,000 individuals aged 55 and older. Over a follow-up period averaging 12.5 years, 749 participants experienced strokes.

Machine learning algorithms helped detect vascular fingerprints in the retinas of these individuals, allowing for a deeper understanding of how subtle changes in blood vessels correlate with stroke risk.

By analysing retinal images and identifying patterns invisible to the human eye, AI revealed that certain features of retinal blood vessels, such as increased arc length and alterations in diameter, were associated with a 9.8–19.5% higher likelihood of stroke.

“Our findings indicate that this association is mainly due to arterial density parameters. Pathologically, this could result from compromised oxygen and nutrient supply.” The team says in its published paper.

This AI-powered approach demonstrated predictive accuracy comparable to traditional methods that assess stroke risk factors, such as age, hypertension, and cholesterol levels. The findings suggest that eye scans could complement existing diagnostic tools, offering a less invasive and more accessible way to assess stroke risk.

With strokes being a leading cause of disability and death worldwide, early detection is crucial. The study’s authors highlight the potential of retinal scans to become a practical tool in primary healthcare settings, “particularly for primary healthcare and low-resource settings.” 

By identifying at-risk individuals early, medical professionals could intervene with lifestyle changes or medical treatments to reduce the likelihood of stroke. The retina’s accessibility and its vascular similarity to the brain make it an ideal candidate for this type of preventive healthcare.