A simple eye scan, commonly performed during routine eye exams, could soon serve as a powerful tool to assess an individual’s risk of heart attack or stroke over the next ten years. Recent research employs artificial intelligence (AI) to analyze digital photographs of the retina—the back of the eye—providing rapid and personalized cardiovascular risk evaluations.
Researchers at the University of Dundee developed an AI algorithm capable of detecting subtle changes in retinal blood vessels that reflect broader vascular health. Since individuals with type 2 diabetes routinely undergo retinal screening to monitor for diabetic retinopathy, this existing testing environment offered an ideal framework for the study.
Dr. Ify Mordi, a British Heart Foundation research fellow and cardiologist at the University of Dundee, explained, “The eyes truly are a window to the heart. When blood vessels in the retina show signs of damage or narrowing, it often mirrors similar changes in vessels supplying the heart, increasing the risk of heart attack or stroke.”
READ MORE: Nine Months of M50 Flooding Repairs to Begin with Overnight Closures and Traffic Delays
READ MORE: Dentist Reveals the Best Time to Brush Your Teeth for Optimal Oral Health
The AI system was trained on 4,200 retinal images, learning to identify markers such as vessel constriction or blockages, as well as variations in vessel size and positioning. Impressively, the technology predicted major cardiovascular events with 70% accuracy within a ten-year timeframe.
“This one-stop retinal scan takes less than a minute and could complement traditional risk assessments like blood pressure and cholesterol checks,” said Dr. Mordi. “It opens the door to earlier lifestyle or medical interventions for those at risk.”
Professor Bryan Williams, Chief Scientific and Medical Officer at the British Heart Foundation, highlighted the promise of such innovations: “Improving accuracy in detecting heart attack and stroke risk is critical to prevention efforts. Integrating retinal imaging with standard screenings could significantly enhance prediction and align with our goal to prevent 125,000 heart attacks and strokes in the UK by 2035.”
While encouraging, experts caution that further validation is necessary to confirm the AI’s predictive reliability and to explore how retinal scans can be integrated into routine clinical care.
Dr. Clare Jonas, research lead at the Stroke Association, added, “AI-based analysis of health data through screening technologies like this eye scan holds great potential for early detection, especially in high-risk groups like people with diabetes. Advances such as these could empower individuals to take timely action to prevent stroke.”
She also noted the importance of addressing practical considerations before widespread clinical adoption and expressed hope that future healthcare plans will embrace such technological progress to reduce stroke incidence.