Key Takeaways:
- A fully autonomous AI system can predict cardiovascular disease risk by analyzing retinal blood vessels, detecting signs like hypertension non-invasively.
- The AI was trained on 70,000 retinal images from diverse populations, enabling it to identify vascular markers linked to heart disease.
- In trials, the AI demonstrated superior accuracy in detecting cardiovascular risk compared to human clinicians.
- Retinal angiography uses a contrast medium to map blood vessels, offering a quick, painless method for early disease detection.
- This technology could revolutionize preventive care, enabling widespread, low-cost cardiovascular screenings during routine eye exams.
By analyzing the retinal blood vessels, a fully autonomous artificial intelligence is able to detect potential cardiovascular diseases.
Artificial Intelligence
Our eyes can tell us much more about our health than we think. An international collaboration of researchers has developed an autonomous artificial intelligence that is able to predict the risk of cardiovascular disease in patients through the blood vessels of the retina. The results of the study were published in the journal Nature Biomedical Engineering.
An autonomous artificial intelligence
The researchers came up with the idea of using angiography to illuminate the blood vessels in the retina. In this method, medical imaging software is used, while the patient is administered a slightly radioactive contrast medium, which then attaches itself to the target organ or tissue. Using the radioactivity of the contrast medium and the rays from the device used to transcribe the information, it is then possible to visually map the area.
In the case of retinal angiography, the aim is to detect hypertension in the blood vessels, which would be a sign of cardiovascular disease. Although it has long been possible to detect the risk of cardiovascular disease in our retinas, the novelty here is that, thanks to artificial intelligence, human intervention will no longer be necessary.
Better results than doctors
To this end, the researchers trained the artificial intelligence by using 70,000 images of people from different backgrounds, so that it could understand on its own whether or not they were at risk of cardiovascular disease. The images focused on common markers that can be identified in the retinal vascular system.
To better assess the accuracy of the artificial intelligence’s recognition capabilities, the researchers then subjected the computer to tests to compare these results with those of real doctors. Without surprise, the recognition abilities of this artificial intelligence surpassed those of real doctors.
Related Reading:
An Artificial Intelligence Capable of Translating Thoughts Into Words Is Now a Reality
FAQs:
How does the AI detect cardiovascular risk from the retina?
The AI analyzes retinal blood vessel patterns (e.g., narrowing, leaks) that correlate with hypertension and systemic vascular damage, which are early indicators of heart disease.
Why use the retina for cardiovascular screening?
Retinal vessels mirror systemic blood vessel health, and changes in their structure (e.g., thickness, branching) reflect conditions like atherosclerosis or high blood pressure.
How accurate is this AI compared to doctors?
In trials, the AI surpassed clinicians in identifying at-risk patients, reducing human error and variability in interpreting retinal scans.
When will this technology be available clinically?
Pending regulatory approvals and further validation, it could integrate into eye exams within 3–5 years, offering preventive screenings alongside vision checks.
What are the benefits over traditional methods?
Non-invasive, fast, and cost-effective compared to MRI or blood tests. Ideal for early detection in asymptomatic individuals.
References
Cheung, C.Y., Xu, D., Cheng, CY. et al. A deep-learning system for the assessment of cardiovascular disease risk via the measurement of retinal-vessel calibre. Nat Biomed Eng 5, 498–508 (2021). https://doi.org/10.1038/s41551-020-00626-4
