How Artificial Intelligence Interprets 'Eyes Are the Window to the Soul'

​Software developed using machine learning can predict someone’s risk of heart disease in less than a minute by analyzing the veins and arteries in their eyes.

The new study was published in the British Journal of Ophthalmology. If the findings are confirmed in future clinical trials, it could pave the way for rapid, inexpensive cardiovascular screening. These screenings can let people know their risk for stroke and heart disease without requiring blood tests or even measuring blood pressure.

Experts say the research is promising but requires further research before it can become a viable diagnostic tool.

"This artificial intelligence tool can let people know their risk level in 60 seconds or less," Alicja Rudnicka, the study's lead author, told The Guardian. It was found that the prediction results of this tool are as accurate as the results of currently commonly used testing methods.

“The eyes can serve as windows to the rest of the body.”

The software works by analyzing the network of blood vessels within the retina. It measures the total area covered by these arteries and veins, as well as their width and "tortuosity" (how curved they are). All of these factors are affected by an individual's heart health, so the software can predict a subject's risk of heart disease simply by looking at non-invasive snapshots of their eyes.

The researchers named their software QUARTZ (an original acronym derived from the phrase "Quantitative Analysis of Retinal Vessel Topology and Size").

We often say that "the eyes are the windows to the soul", and more and more knowledge shows that the eyes can be used as a diagnostic window to other parts of the body. Doctors have known for more than a hundred years that signs of diabetes and high blood pressure can be seen through the eyes. But the problem is that in the past, manual assessment was based on the experience of medical experts, which increased uncertainty and misjudgment. This challenge can now be overcome using machine learning.

Using artificial intelligence to diagnose disease through eye scans has proven to be one of the fastest growing areas of machine learning medicine. The first AI-powered diagnostic device approved by the FDA was used to screen for eye disease, and research suggests AI could detect a range of conditions in this way, from diabetic retinopathy to Alzheimer's disease. These application tools are at different stages of development, but questions remain about their diagnostic reliability and generalizability.

For example, the latest study, by a team at St. George's College, University of London, only tested eye scans on white patients. The team obtained test data from the UK Biobank, which happens to be 94.6% white (reflecting the UK's own demographics, including the age range of patients in the biobank). This bias must be balanced in the future to ensure that any diagnostic tool is equally accurate across ethnic groups.

The researchers compared the results of their software QUARTZ with 10-year risk predictions produced by the standard Framingham Risk Score test (FRS). They found that the two methods had "comparable performance."

Experts say the biggest challenge is taking this kind of work from "coding to clinical." For example, who can turn this kind of research into a formal diagnostic tool? Is it the UK's National Health Service (NHS) or a company spun out of a university? What will regulators require before approving the use of the software? Performance level? There is still a long way to go from research to product practicality (commercialization). ​

The above is the detailed content of How Artificial Intelligence Interprets 'Eyes Are the Window to the Soul'. For more information, please follow other related articles on the PHP Chinese website!