- A new study, published in the British Journal of Ophthalmology, opens the way to the development of rapid and inexpensive cardiovascular screenings if its results are confirmed in future clinical trials.
- The study found that the predictions were as accurate as those made by modern tests.
- It measures the total area covered by these arteries and veins and their width and degree of curviness.
If the findings of a recent study are supported by subsequent clinical trials, they could pave the way for the creation of quick and affordable cardiovascular tests. The study was published in the British Journal of Ophthalmology. Without the need for blood testing or even blood pressure readings, these screenings will give individuals the opportunity to learn about their risk of heart attack and stroke as reported by Mezha.
Accurate results
According to the lead study author Alicja Rudnicka, “This AI tool could let someone know in 60 seconds or less their level of risk.” According to the study, the predictions were just as accurate as the results of more recent studies.
The programme analyses the retina’s blood vessel network to perform its function. It determines the entire area that these arteries and veins encircle, in addition to their width and degree of curvature. The software can forecast a person’s risk of heart disease based solely on looking at a non-invasive image of their eyes because all of these characteristics have an impact on a person’s heart health.
Machine learning application
According to Pearse Keane, an ophthalmologist and AI analyst who was not involved in the study, “The study adds to a growing body of understanding that the eye can be utilised as a window to the rest of the body.” “Doctors have known for more than a century that diabetes and high blood pressure symptoms can be seen in the eyes. However, the issue was with the human specialists’ manual delineation of the vessels. Keanea claims that this difficulty can be solved by the application of machine learning.”
Concerns remain
One of the fastest-growing fields in machine learning medicine is the use of AI to identify diseases based on eye scans. Studies suggest that AI may detect a variety of illnesses this way, from diabetic retinopathy to Alzheimer’s disease (an area of Keane’s own research). The first-ever FDA-approved AI diagnostic gadget was used to check for eye diseases. Although the tools used to apply these discoveries are in various phases of development, concerns remain regarding the accuracy and applicability of their diagnoses.
UK demographics only?
Only scans of white patients’ eyes were used in this most recent investigation, which was conducted by a team from St. George’s University of London. The team used the UK Biobank, a 94.6% white database, to get data for testing (which includes UK demographics for the age range of patients included in the Biobank). To ensure that any diagnostic tool is similarly accurate across ethnic groups in the future, such biases must be balanced.
Results contrast
The results of the researchers’ software, QUARTZ (an original acronym derived from the phrase “QUantitative Analysis of Retinal vessels Topology and siZe,” were contrasted with 10-year risk projections generated using the widely used Framingham Risk Score test (FRS). They discovered that the two techniques “perform comparably.”
Did you subscribe to our daily Newsletter?
It’s Free! Click here to Subscribe
Source: Mezha
