Using different types of health data - from body measurements to genetics - the authors have built on evidence that there is a potential future where simple, high street eye scans could be used as a measure of how the rest of the body is functioning.
Using “UK Biobank” data, a UK volunteer-based study, the researchers developed an AI tool called “Ret-AAE” to explore the links between the eye and disease risk, blood test results, and the appearance and function of different organs, across over 68,000 people.
The study showed that the associations between the eye and body are incredibly broad, identifying that the appearance of the eye is linked to risk of heart failure, high blood pressure, heart attack, Parkinsons disease, dementia and more.
Two types of scan were used in the study - 3D scans of the inner lining of the eye called “optical coherence tomography” (OCT), and simple photographs of the back of the eye called “colour fundus photographs”.
Both scan types are widely available at optometrists across the UK, with several million scans already captured every year by high street providers – making them a highly accessible health marker.
The team showed the two scan types might reveal complementary signals about our future health, with OCT more strongly linked to neurological traits and CFP having broader associations with cardiovascular traits.
Further analysis showed that the eye contains several signals that are captured by AI systems and reveal information about health – including the appearance of blood vessels and the nerves which connect the eye to the brain.
Some of the patterns the system picked up were caused by cataracts or natural differences in eye colour, showing that researchers might need to account for age and ethnicity when analysing the images.
One of the key contributions of the study is the work around the biological pathways that might link the eye to disease in the brain, blood vessels and heart.
Genetic analyses showed that eye features are linked to genes involved in neurodegenerative disease pathways, including those related to Parkinson’s disease, dementia, and broader neurodegeneration.
Physiological analyses linked eye features to blood pressure, blood vessel stiffness, and the function of the heart.
‘Radiomic analyses’ - which turn medical images into measurable data - showed associations between eye features and the size of the brain, as well as tiny changes in the brain’s tissue structure detected using MRI scans.
By studying the tiny molecules in the blood, the researchers also found several connections between features in the eye and fat-related molecules in the body, which may link the eye to general health.
Lead author Dr Tom Julian, Medical Research Council Clinical Research Training Fellow, is an eye doctor and researcher at the University of Manchester and Manchester Royal Eye Hospital, part of Manchester University NHS Foundation Trust.
He said: “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.
“Our 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, from The University of Manchester, who oversaw this interdisciplinary work alongside Professor Alejandro Frangi FREng, also from the University, said: “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.”
Professor Frangi is also a RAEng Chair and Digital Infrastructure Programme Co-Lead at the National Institute for Health and Care Research (NIHR) Manchester Biomedical Research Centre’.
He said: “Using scans available on every high street, an eye test may become much more than a way to check your glasses prescription.”
“This work shows the interdisciplinary work ongoing in Manchester thanks to generous support by various translational structures, including the NIHR Manchester Biomedical Research Centre, and the BHF Manchester Centre of Research Excellence, among other funders”.
Multi-omic analysis of deep learning-derived phenotypes links ophthalmic imaging to cardiovascular and neurological traits is published in Nature Cardiovascular Research on Tuesday 16 June at 4pm (UK time).
- Funders included: the Medical Research Council; the Wellcome Trust; the British Heart Foundation; the Royal Academy of Engineering; and the National Institute for Health and Care Research (NIHR) Manchester Biomedical Research Centre.