A new AI-enabled stethoscope, developed by California-based Eko Health and tested by researchers from Imperial College London and Imperial College Healthcare NHS Trust, has been shown to detect three serious heart conditions, namely heart failure, atrial fibrillation and heart valve disease in as little as 15 seconds.
The device combines the traditional listening function with a built-in microphone and electrocardiogram (ECG) capability to pick up subtle changes in heart rhythm and blood flow that are often inaudible to the human ear.
Data collected from the device is securely uploaded to the cloud, where algorithms trained on tens of thousands of patient records analyse it. A test result is then sent directly to a smartphone, allowing general practitioners to act immediately if a patient is flagged as at risk.
The technology was trialled in the UK through the TRICORDER study, involving over 12,700 patients across 200 GP surgeries covering a population of around 1.5 million. Patients presenting with symptoms such as breathlessness, fatigue, or leg swelling were assessed with the AI stethoscope, and their diagnoses were compared with those of patients seen at practices not using the tool.
The results, presented at the European Society of Cardiology annual congress in Madrid, were significant. Patients assessed with the AI device were 2.3 times more likely to be diagnosed with heart failure, 3.5 times more likely to be diagnosed with atrial fibrillation, and nearly twice as likely to be diagnosed with heart valve disease within a year. Researchers say this earlier detection could allow doctors to prescribe lifesaving treatments sooner, rather than waiting until patients are admitted to a hospital in an emergency.
However, the study also highlighted challenges. Around 70% of GP practices given the devices stopped using them regularly after a year, with researchers pointing to the need for better integration into clinical workflows and more training. The device was also prone to false positives: in some cases, patients were flagged as being at risk when follow-up tests showed no underlying condition. Experts caution that while this could lead to unnecessary investigations, it is still preferable to missing serious cases that might otherwise go undiagnosed.
The AI stethoscope, about the size of a playing card, is not intended for routine checks in healthy patients but could prove a valuable tool in primary care settings where early diagnosis is crucial. Researchers believe that if adoption hurdles can be addressed, this technology could mark the biggest leap forward for the stethoscope in more than two centuries.