AI-powered eye scans may help detect heart and brain disease risk years earlier. Researchers at The University of Manchester have developed an artificial intelligence tool that links routine eye scans to markers of cardiovascular and neurological health, raising the possibility of future disease screening through standard optometry tests.
Routine eye tests could one day help identify people at risk of heart failure, dementia, Parkinson’s disease and other serious conditions before symptoms emerge, according to a new study led by researchers at The University of Manchester.
The findings suggest that artificial intelligence (AI) can extract signals from commonly used eye scans to reveal information about cardiovascular and neurological health. If validated in future research, the approach could expand the role of routine eye examinations beyond vision care and into preventive health screening.
The study, published in Nature Cardiovascular Research, analysed health and imaging data from more than 68,000 participants in the UK Biobank, a large volunteer-based research programme in the United Kingdom.
How the researchers studied the eye-body connection
The research team developed an AI tool called Ret-AAE to investigate links between eye characteristics and a wide range of health measures, including disease risk, blood test results, genetics and organ function.
Researchers used two types of ophthalmic imaging:
- Optical coherence tomography (OCT): A 3D scan of the inner lining of the eye.
- Colour fundus photography (CFP): A photograph of the back of the eye.
Both technologies are already widely available in optometry settings, making them attractive candidates for future large-scale screening programmes.
The analysis found that eye appearance was associated with risks of heart failure, high blood pressure, heart attack, Parkinson’s disease, dementia and other health conditions.
Different scans reveal different health signals
The study found that the two imaging methods appeared to provide complementary information about health.
OCT scans showed stronger links with neurological traits, while colour fundus photographs demonstrated broader associations with cardiovascular traits.
Further analysis suggested that AI systems detect several biological signals within eye images. These include characteristics of blood vessels and the nerves that connect the eye to the brain.
Researchers also found that some image patterns were influenced by factors such as cataracts and natural variations in eye colour. The findings indicate that age and ethnicity may need to be considered when interpreting eye scans for health assessment.
Genetic and biological pathways identified
A key aspect of the study was its effort to understand the biological mechanisms connecting eye features with disease processes elsewhere in the body.
Genetic analyses linked eye characteristics to genes involved in neurodegenerative disease pathways, including those associated with Parkinson’s disease, dementia and broader neurodegeneration.
Physiological analyses identified associations between eye features and blood pressure, blood vessel stiffness and heart function.
The researchers also used radiomic analysis, a method that converts medical images into measurable data. This analysis revealed links between eye features, brain size and subtle changes in brain tissue structure detected through magnetic resonance imaging (MRI).
In addition, the team identified connections between eye characteristics and fat-related molecules circulating in the blood, suggesting potential links between eye health markers and broader metabolic processes.
Researchers see potential for preventive screening
Dr Tom Julian, Medical Research Council Clinical Research Training Fellow, eye doctor and researcher at The University of Manchester and Manchester Royal Eye Hospital, said the findings demonstrate that the eye can provide insights into overall health.
“Our findings show that the eye can reveal a remarkably broad picture of whole-body health, offering a way to identify those at risk of heart and brain disease before they occur,” Julian said.
He added that the study advances the use of deep-learning-derived eye traits in large-scale biomedical research.
Dr Panos Sergouniotis, Wellcome Clinician Scientist, Senior Lecturer and Honorary Consultant at The University of Manchester, said more research is needed before such tests become part of routine community healthcare.
“While more work is needed before these tests could arrive on the high street, we hope and believe that routine eye tests will one day be used as part of health screening for disease prevention,” Sergouniotis said.
Professor Alejandro Frangi, who co-led the work and serves as a Royal Academy of Engineering Chair and Digital Infrastructure Programme Co-Lead at the National Institute for Health and Care Research (NIHR) Manchester Biomedical Research Centre, said the findings point to a broader future role for ophthalmic imaging.
“Using scans available on every high street, an eye test may become much more than a way to check your glasses prescription,” he said.
What the findings mean
Artificial intelligence uses computer systems to identify patterns in large datasets that may not be visible through conventional analysis. In this study, AI was used to analyse eye images and identify associations with cardiovascular and neurological traits.
The research does not mean that eye scans can currently diagnose heart disease, dementia or Parkinson’s disease. However, the findings suggest that ophthalmic imaging may eventually become a non-invasive tool for identifying people who could benefit from further medical evaluation.
As healthcare systems increasingly focus on disease prevention and early detection, researchers believe the eye could emerge as a valuable window into whole-body health.
