A standard digital photograph of the retina may soon become a powerful tool in predicting a person’s risk of heart attack or stroke within the next ten years. A team of international researchers from the University of Dundee has developed an AI-driven system that can analyze these images and deliver an individualized cardiovascular risk profile in mere seconds. The findings were published in the journal Cardiovascular Diabetology.
The algorithm focuses on identifying subtle vascular changes—such as narrowing, blockages, or structural damage—within the retina that mirror the condition of the entire cardiovascular system. In a study involving over 1,200 participants, those flagged by the AI as having the most dramatic rise in risk over a three-year period were found to be 54% more likely to suffer cardiovascular events.
“This approach could transform routine eye exams into lifesaving screening tools,” the researchers noted.
The predictive accuracy of the AI model reached 70% using only retinal images, and 73% when combined with traditional risk assessments and genetic testing. Experts believe that this innovation could soon be integrated into regular ophthalmologic check-ups, making early intervention more accessible.
According to the study’s authors, this technology enables earlier identification of at-risk individuals and supports timely preventive action—from lifestyle changes to medical treatment—with the goal of significantly reducing the incidence of heart attacks and strokes in the future.
In a related finding, scientists previously discovered that drinking plain water regularly was associated with a reduced risk of stroke, further emphasizing the value of simple preventive habits.
