AI stethoscope can help spot life-threatening heart disease years earlier, study finds

A stethoscope that uses artificial intelligence could help doctors detect serious heart valve disease years earlier, potentially saving thousands of lives, a new study suggests.

It is estimated that 41 million people worldwide, including 1.5 million people in the UK, live with a type of heart valve disease, which can lead to heart failure, hospital admissions and death.

Early diagnosis is vital for successful treatment, but the condition can be symptom-free in its early stages before causing dizziness, shortness of breath and heart palpitations, which can be confused with other conditions, meaning some patients do not get a diagnosis until the disease is advanced.

Currently, diagnosis of valve disease relies on echocardiography, a type of ultrasound scan that is expensive and time-consuming. While doctors do listen to the heart using a stethoscope, this is not routinely done in short GP appointments, and is known to miss many cases.

But the new technology that works with digital stethoscopes was found to outperform GPs at detecting valve disease, and could be used as a rapid screening tool.

“Valve disease is a silent epidemic,” said Professor Anurag Agarwal from Cambridge’s department of engineering, who led the research. “An estimated 300,000 people in the UK have severe aortic stenosis alone, and around a third don’t know it. By the time symptoms appear, outcomes can be worse than for many cancers.”

For the study published in the journal npj Cardiovascular Health, researchers analysed heart sounds from nearly 1,800 patients using an AI algorithm trained to recognise valve disease.

The AI was found to correctly identify 98 per cent of patients with severe aortic stenosis, the most common form of valve disease requiring surgery, and 94 per cent of those with severe mitral regurgitation, where the heart valve does not fully close and blood leaks backwards across the valve.

Rather than training the algorithm to recognise heart murmurs, the researchers trained it directly on echocardiogram results. They say this allowed the system to learn subtle acoustic patterns that humans might miss, including cases with no obvious murmur.

When tested against 14 GPs who listened to the same recordings, the algorithm was found to outperform every single one, delivering reliable results.

The researchers say the system was designed to minimise false alarms, reducing the risk of overwhelming already-stretched echocardiography services. They add that the technology is not intended to replace doctors, but could be a useful screening tool, helping doctors decide which patients should be referred for further investigation and treatment.

Further trials, carried out in real-world GP settings with a diverse group of patients, will be needed before the device can be rolled out, according to the researchers.

However, they say that AI could help address growing pressures on the health service caused by an ageing population.

“Valve disease is treatable. We can repair or replace damaged valves and give people many more years of healthy life,” said co-author Professor Rick Steeds, from University Hospitals Birmingham. “But timing is everything. Simple, scalable screening tools like this could make a real difference by finding patients before irreversible damage occurs.”

AI is already being trialled in healthcare settings to diagnose cancer and lung conditions.

There is currently an ongoing trial of 700,000 women in the UK testing how AI tools can be used to catch breast cancer earlier. Previous studies have found that AI helps to reduce the rate of breast cancer diagnosis by 12 per cent in the years following a screening.

AI and robotics are also being used at Guy’s and St Thomas’s NHS Foundation Trust to identify abnormal spots or nodules. So far, 300 patients have undergone the robotic biopsy procedure, with 215 of them going on to receive cancer treatment.

Robotics are also being used to perform life-saving bowel cancer surgery, offering a less invasive option with a faster recovery time.