Artificial intelligence can tell if you’ve got heart problems simply by the sound of your voice

WASHINGTON — Advanced AI technology can now calculate a person’s risk of suffering a heart attack just by listening to their voice, a new study says. Researchers at the Mayo Clinic say people who have a high voice biomarker score are more than twice as likely to suffer major heart problems related to clogged arteries.

Coronary heart disease is the most common cause of heart attack and is one of the leading causes of death for both men and women worldwide. It occurs when the heart’s blood supply is blocked or interrupted by a build-up of plaque in the arteries. Now, scientists have developed a “powerful screening tool” dubbed the Vocalis Health algorithm to help identify those who are particularly at risk and may need close monitoring.

“Telemedicine is non-invasive, cost-effective and efficient and has become increasingly important during the pandemic,” says lead author Dr. Jaskanwal Sara in a media release. “We’re not suggesting that voice analysis technology would replace doctors or replace existing methods of health care delivery, but we think there’s a huge opportunity for voice technology to act as an adjunct to existing strategies. Providing a voice sample is very intuitive and even enjoyable for patients, and it could become a scalable means for us to enhance patient management.”

6 voice features linked to heart problems

The researchers recruited 108 patients referred for an X-ray to assess the condition of their coronary arteries. The participants had to record three 30-second voice samples using the Vocalis Health smartphone application. During the first two recordings, each person read from a prepared text before speaking freely about a positive experience. For the third recording, the group did the same but spoke about a negative experience. The Vocalis Health algorithm then analyzed all of the samples.

Researchers trained the AI-based system using more than 10,000 voice recordings from Israel however, they add it is not yet ready for clinical use. It can pick up on 80 different voice features, like their frequency, amplitude, pitch, and cadence.

The results linked six features to coronary artery disease, before combining all of the readings into a single score, expressed as a number between minus one and one. “We can’t hear these particular features ourselves,” Sara says. “This technology is using machine learning to quantify something that isn’t easily quantifiable for us using our human brains and our human ears.”

A high voice biomarker score nearly triples heart disease risk

Over the next two years, 58.3 percent of the participants with a high voice biomarker score visited a hospital for chest pains or major heart problems including heart attacks. In contrast, only 30.6 percent of those with a low voice biomarker score ended up in the hospital. Overall, those with a high voice score were 2.6 times more likely to suffer major problems associated with coronary artery disease (CAD).

These patients were also three times more likely to show evidence of plaque build-up in medical tests compared with those who had a low score. While the researchers do not know exactly why a person’s voice may reveal their level of heart health, they believe the autonomic nervous system could be an important piece of the puzzle. This part of the nervous system regulates bodily functions which are not under a person’s conscious control including many aspects of their voice and heart, like blood pressure and heart rate.

Similar technology has been tested on a range of disorders, including Parkinson’s disease, Alzheimer’s disease, and COVID-19.

Researchers are presenting their findings at the American College of Cardiology’s 71st Annual Scientific Session.

South West News Service writer Tom Campbell contributed to this report.