A new AI system that evaluates cardiovascular risk using eye images from eye exams shows a strong correlation with more conventional assessments.
This finding comes from a study presented at the American College of Cardiology’s Annual Scientific Session.
The researchers suggest that using AI to screen for heart disease risk during routine eye exams could raise awareness of heart disease risk among more people and facilitate referrals for preventive care.
Michael V McConnell, clinical professor of medicine at Stanford University in Stanford, California, and the study’s lead author, said: ‘The awareness that someone might be at risk is really one of the key missing pieces. The image of the back of your eye has a wealth of health information. We can analyse these images with AI to help people become aware of their risk and have the opportunity to get guideline-based evaluation and preventive therapy.’
McConnell serves as chief health officer at Toku, the company that developed the AI system used in the study.
The system, called CLAiR, received a ‘breakthrough device’ designation from the US FDA. The results of this first prospective evaluation of CLAiR in the US will support the FDA submission.
In this study, 1 in 4 participants were found to have an elevated risk of heart disease based on standard cardiovascular risk assessments, including blood pressure and cholesterol screening.
The AI-based method that analysed blood vessels at the back of the eye using retinal images taken during the visit largely aligned with this determination, identifying at-risk participants with a sensitivity of 91.1% and a specificity of 86.2%.
McConnell said: ‘Even just a standard retinal photo provides high-resolution imaging of your blood vessels – it’s a literal window into vascular tissue.’
Previous studies have shown that eye images can be used to assess conditions such as diabetes, but most methods have relied on interpretation by experts.
With CLAiR, developers sought to demonstrate how this approach can be scaled for clinical implementation by using AI to automate image analysis.
The AI system was trained to recognise patterns in blood vessel appearance associated with the development of heart disease.
Based on the results, researchers said the AI system shows promise as a non-invasive screening method.
However, more work is needed to facilitate referrals of at-risk patients for cardiovascular evaluation and treatment in primary care after retinal image screening.
McConnell said: ‘This approach would not replace the standard cardiovascular risk evaluation, but it’s a potential way to bring greater awareness, especially for people who should be on preventive care, but who have not yet had a thorough evaluation. For patients to benefit, we need to implement clear pathways to connect your elevated risk from your eye exam to help you see your clinician and ultimately get guideline-based preventive therapy.’
Overall, 94% of the images acquired in the study were usable by the AI system, providing evidence that the approach works well across cameras used in different clinics.
Retinal imaging takes about five minutes, and the CLAiR algorithm returns results in about 30 seconds, suggesting that implementing this approach would add little time to clinical workflows.
