AI stethoscope flagged heart conditions twice as often as doctors in trials

Researchers have developed an artificial intelligence (AI)- powered stethoscope that can spot three major heart conditions in just 15 seconds, potentially saving thousands of lives through earlier diagnosis.

The device, developed by Imperial College London scientists, detected heart failure more than twice as often as traditional methods during trials involving over 12,000 patients. It also found abnormal heart rhythms 3.5 times more frequently and heart valve disease nearly twice as often.

“It is incredible that a smart stethoscope can be used for a 15-second examination, and then AI can quickly deliver a test result indicating whether someone has heart failure, atrial fibrillation or heart valve disease,” said Dr. Patrik Bächtiger, who led the research team.

Traditional stethoscopes rely on doctors’ ears to detect problems. This new version utilizes AI trained on data from tens of thousands of patients to detect subtle signs that humans may miss.

The playing card-sized device sits on a patient’s chest to record heart sounds and electrical signals. Information is transmitted to cloud servers, where AI algorithms analyze the data. Results return to doctors’ smartphones within seconds.

Source: Imperial College London

During the TRICORDER study, researchers compared 96 general practitioner (GP) surgeries using the AI tool with 109 practices providing standard care. Patients examined with the smart stethoscope were 2.33 times more likely to receive heart failure diagnoses within 12 months.

Heart failure affects about 6.7 million Americans and often goes undetected until patients arrive at emergency rooms in crisis. The condition causes breathlessness, fatigue, and swelling in the legs and feet.

“Most people with heart failure are only diagnosed when they arrive in A&E seriously ill,” explained Dr. Mihir Kelshiker, another team member. “This trial shows that AI-enabled stethoscopes could change that.”

The device also excels at finding atrial fibrillation. This irregular heartbeat increases stroke risk but often causes no symptoms. Early detection allows doctors to prescribe blood thinners that prevent strokes.

California company Eko Health manufactured the stethoscopes used in the study. The research team presented their findings at the European Society of Cardiology Congress in Madrid in late August.

However, the technology carries risks. Two-thirds of patients flagged for heart failure turned out not to have the condition after further testing. Researchers emphasized that the tool should only be used on patients exhibiting symptoms, not for routine checkups on healthy individuals.

“Given an earlier diagnosis, people can access the treatment they need to help them live well for longer,” said Dr. Sonya Babu-Narayan from the British Heart Foundation, which helped fund the research.

The team plans to expand the program across Wales, Sussex, and other UK regions. Each patient diagnosed early could save the health system £2,400 by avoiding emergency visits.

The breakthrough demonstrates how AI can enhance rather than replace human doctors. While AI spots patterns in data, physicians still make final treatment decisions based on their clinical judgment and patient needs.