AI-guided ultrasound screening by lay users for aortic stenosis

What the research shows

The study reported that 1,302 patients were assessed by people with no clinical or ultrasound experience. After 4 hours of training, these lay users performed ultrasound guided by artificial intelligence to identify moderate or severe aortic stenosis, a narrowing of the valve that controls blood flow from the heart into the aorta. According to the report, the AI alone had a sensitivity of 93.5% and a specificity of 96.4%, meaning it correctly flagged most people who had the disease and correctly cleared most people who did not.

Sensitivity is the share of people with the disease that the method detects, and specificity is the share of people without it that the method correctly rules out. Because this describes a study evaluating the method, the result shows performance in the group studied, but it does not prove the same screening would work the same way in every health service. The authors' conclusion, as presented, is that AI-guided screening may expand detection of heart valve problems beyond specialist teams, which still needs confirmation from other studies.

What it means

The finding suggests artificial intelligence may help people with limited training screen for a heart valve disease. It is still unclear how well this performance holds up in other settings, with other equipment and other patient groups, or what effect it would have on diagnosis and treatment.

How to read it with care

This result does not replace a medical consultation. If you have heart symptoms such as shortness of breath, chest pain or fainting, see a doctor. Any decision about tests, diet changes, supplements or treatment should be made with a doctor or another qualified professional.