Researchers have developed an new AI powered eye exam that can determine the likelihood of that person developing heart disease in under a minute. The discovery paves the way for a very accurate test, that does not require any intrusive procedures, and does not even require a visit to the doctors office.
By analyzing photographs of the retina with an artificial intelligence tool, doctors may in the future be able to estimate a person’s risk of developing heart disease. Based on a large research study, the new test may make it possible for ophthalmologists and other general physicians to offer cardiovascular screening using a retinal camera. It is being studied as an addition to standard care and does not replace blood tests, stress tests or blood pressure checks.
The AI algorithm uses imaging of the veins and arteries in the retina to define a persons risk of getting cardiovascular disease, cardiovascular mortality,heart attacks or brain injuries. Using this advanced technique, a patients’s degree of risk can be determined in less than one minute. If a patient’s risk is found to be higher than expected, their doctor could discuss proven prevention steps such as lifestyle changes, blood pressure control and statins; treatments that aim to promote angiogenesis are not a recognised prevention option. Researchers hope tools like this could help identify people at risk earlier, but they still need to be tested in clinical practice.
Identifying Circulatory system diseases
Diseases of the heart and circulatory system are among the leading causes of illness and mortality in the world. These conditions include coronary heart disease, congestive heart failure, stroke, peripheral artery disease and other vascular diseases. Lung diseases such as COPD and IPF are separate respiratory conditions, although they often occur alongside heart disease. Cardiovascular disease is the leading cause of death in the world and is estimated to be responsible for about 1 in 3 deaths each year.
The test is not perfect yet, however. For some “high risk” patients the indicated risk, isnt always able to accurately identify those who will go on to develop heart disease later in life. Another automated artificial intelligence-enabled device called “Quartz” in being developed by researchers to evaluate the potential of retinal vasculature imaging, in conjunction with known risk factors, to better forecast vascular health and mortality rates.
The improved model was trained on data from patients from age 35 to 70 years old. When developing medical prediction models for stroke, heart disease, and mortality from circulatory disease, researchers look to measure the architecture of the arteries and veins in the retina, the area of the vessels and the degree to which they were curved. After the initial scan, they researchers used the AI models to analyze the retinal pictures of of over 9,500 patients who participated in the research. The participants’ assigned risk was compared to established cardiovascular risk prediction models.
The overall health of each participant was monitored for a period of five to seven years, on average. Once the trial was over researchers discovered that the breadth, curviness, and width variation of veins and arteries in the retina were strong predictors of mortality from cardiovascular illness. This was the case, particularly in men, and variations in artery area and breadth and vein curviness and width contributed to risk prediction in females. The artificial intelligence model also used the individuals’ medical history data, including smoking history, medications used to manage high blood pressure, and prior comorbidities.
The findings showed that the retina data was strongly related to increased risk of cardiovascular illness and strokes, and that its prediction ability was comparable to that of an established clinical risk score, the Framingham risk score.
Benefits of using AI for Medical Screening
AI-enabled vasculometry risk prediction models are entirely automated, very low-cost, and non-invasive. Because it relies on retinal photographs, the approach has the potential to reach more people around the world, including places where blood tests are harder to access, although it would still be used alongside standard checks.
The findings lend credence to the results of numerous other research that found the retina to be an informative and possibly game-changing source of data on cardiovascular disease risk in the context of personalized functional medical treatments.
