An AI-powered stethoscope, developed for the TRICORDER study by Imperial College London and Imperial College Healthcare NHS Trust, can detect three serious heart conditions in just 15 seconds.
The AI stethoscope features a single-lead ECG sensor and a microphone array that records phonocardiogram (PCG) waveforms, which are then analyzed by algorithms to identify heart activity and blood flow abnormalities. It connects to a smartphone app via Bluetooth and transmits data to the cloud using Wi-Fi or a cellular network.
Classified as a Class IIa medical device, the AI-driven stethoscope is authorized for regular healthcare use without requiring explicit patient consent. It captures a 15-second recording from the upper left sternal border area on the chest. The algorithms, governed by the UK Medicines and Healthcare Products Regulatory Agency (MHRA), have shown promising and consistent statistical performance validated against international external studies, according to a research paper published in the BMJ Open journal.
Initial studies involving over 200 GP surgeries and more than 1.5 million patients revealed significant improvements in diagnosis rates. Patients examined with the AI stethoscope were approximately 3.5 times more likely to be diagnosed with atrial fibrillation and almost twice as likely to receive a diagnosis of heart valve disease, according to the British Heart Foundation, which partly funded the studies.
The primary objective of this smart stethoscope is to facilitate early detection of heart problems, enabling timely access to life-saving medications and treatment. Currently, heart failure, affecting around one million individuals in the UK, is often detected late, with 70% of cases diagnosed during emergency care.
Professor Nicholas Peters, senior investigator from Imperial College London and consultant cardiologist at Imperial College Healthcare NHS Trust, emphasized the device’s ability to detect three serious heart conditions simultaneously. The stethoscope is manufactured by Eko Health, a California-based company.
The NHS estimates that implementing this tool in primary care could save £2,400 per patient by potentially preventing unplanned A&E visits. The TRICORDER trial program has already enrolled over three million patients, with potential healthcare savings estimated to exceed a hundred million pounds for the government.
The AI-powered stethoscope offers a promising solution for early and accurate detection of heart conditions, potentially leading to improved patient outcomes and significant cost savings for healthcare systems. The device, roughly the size of a playing card, is provided as a replacement stethoscope for routine patient clinical examination to general practitioners (GP) across the country.
“The statistical performance of these three AI algorithms has been shown to be high and consistent against international external validation studies,” the research team stated in their BMJ Open journal publication.
“It is estimated that implementing this tool in primary care could save the NHS £2,400 per patient by eliminating the potential need for an unplanned A&E visit,” claims the UK’s National Health Service (NHS).
