Doctors develop AI stethoscope that can detect major heart conditions in 15 seconds
Researchers at Imperial College London have unveiled an AI‑enhanced stethoscope that can identify heart failure, valve disease and atrial fibrillation within a 15‑second exam. A large UK trial showed the device doubled detection rates for heart failure and tripled detection of atrial fibrillation c…
Doctors at Imperial College London and Imperial College Healthcare NHS Trust have introduced a groundbreaking artificial‑intelligence (AI) stethoscope that can flag three major cardiac conditions in just 15 seconds. By merging traditional auscultation with rapid electrocardiogram (ECG) recording and cloud‑based AI analysis, the device promises to transform early detection of heart failure, heart‑valve disease and atrial fibrillation in primary‑care settings.
How the AI stethoscope works
The new tool, manufactured by California‑based Eko Health, resembles a playing‑card‑sized handheld device. When placed on a patient’s chest, it simultaneously captures the acoustic signature of blood flow and records a brief ECG trace. These data streams are uploaded to a secure cloud platform where proprietary AI algorithms compare subtle waveform patterns against a massive training set. Within seconds the system returns a binary risk flag for each of the three target conditions to a clinician’s smartphone.
Unlike a conventional stethoscope, which relies solely on the clinician’s hearing, the AI model can discern minute variations in heart sounds and electrical activity that are imperceptible to the human ear. This capability enables a rapid, objective assessment that does not require specialist interpretation.
Evidence from a large‑scale UK trial
The technology was evaluated in a prospective study involving roughly 12,000 patients across 200 general‑practice surgeries in England. Participants presented with symptoms such as breathlessness, fatigue or palpitations—clinical clues that often precede serious cardiac disease.
- Patients examined with the AI stethoscope were twice as likely to receive a diagnosis of heart failure compared with those assessed using standard clinical judgment.
- The likelihood of detecting atrial fibrillation rose threefold, a crucial finding given the rhythm disorder’s strong association with stroke.
- Diagnosis of heart‑valve disease increased by nearly 90%, highlighting the device’s sensitivity to murmurs and flow disturbances that may be missed on routine auscultation.
These results suggest that the AI‑enabled exam can uncover hidden disease earlier, potentially allowing timely initiation of guideline‑directed therapies.
Clinical implications and cautions
Lead investigator Dr Patrik Bächtiger emphasized that the stethoscope’s design marks the first major evolution of the instrument in two centuries. "A 15‑second AI‑driven test that instantly reports risk for heart failure, atrial fibrillation or valve disease could reshape primary‑care pathways," he said.
However, the researchers warned of false‑positive risk flags. The device is intended for symptomatic patients where the pre‑test probability of disease is already elevated, not for routine screening of asymptomatic individuals. Over‑diagnosis could lead to unnecessary investigations, anxiety and healthcare costs.
Dr Mihir Kelshiker added that most heart‑failure cases are currently identified only after patients present in emergency departments with advanced symptoms. By equipping general practitioners with a quick, reliable screening tool, the AI stethoscope may shift diagnosis earlier in the disease trajectory.
Funding, partnerships and future rollout
The project received joint funding from the British Heart Foundation, the National Institute for Health and Care Research (NIHR) and the Imperial College National Heart and Lung Institute. Prof Mike Lewis of NIHR highlighted the broader public‑health impact: "Early detection of these three conditions could reduce mortality and healthcare expenditure, especially in ageing populations where cardiovascular disease is a leading cause of death."
Following the positive trial outcomes, the research team plans to seek regulatory approval across Europe and the United States. Integration with electronic health‑record systems and reimbursement pathways will be critical for widespread adoption.
What comes next?
While the AI stethoscope shows promise, further studies are needed to validate its performance in diverse populations and to assess long‑term outcomes such as reduction in hospital admissions and mortality. Ongoing work will also explore expanding the AI model to detect additional cardiac abnormalities, including hypertrophic cardiomyopathy and congenital defects.
If successful, the technology could become a staple of community clinics, enabling clinicians to triage patients more effectively and refer high‑risk individuals for definitive imaging or specialist care without delay.
Why it matters
Early, accurate detection of heart failure, valve disease and atrial fibrillation can dramatically improve patient outcomes and reduce costly emergency admissions.
Key points
- AI‑enhanced stethoscope delivers a 15‑second risk assessment for three major heart conditions
- Large UK trial of ~12,000 patients showed doubled heart‑failure detection and three‑fold rise in atrial‑fibrillation diagnosis
- Device combines acoustic and ECG data, analysed in the cloud by proprietary AI algorithms
- Intended for symptomatic patients; false‑positives remain a concern
- Funding from British Heart Foundation, NIHR and Imperial College supports further development
Frequently asked questions
How quickly can the AI stethoscope provide a result?
The device records heart sounds and a brief ECG, uploads the data to the cloud, and returns a risk flag for each condition within about 15 seconds.
Which heart conditions can the AI stethoscope detect?
It is trained to identify heart failure, heart‑valve disease and atrial fibrillation.
Is the AI stethoscope meant for routine screening of healthy people?
No. The developers recommend its use only for patients who present with symptoms suggestive of cardiac disease, to minimise false‑positive results.
Who manufactured the AI stethoscope?
The hardware is produced by Eko Health, a medical‑device company based in California.
What are the next steps before the device reaches clinics?
The team aims to obtain regulatory clearance, integrate with electronic health‑record systems and conduct further studies to confirm long‑term clinical benefits.
