Experts Find Ways to Detect Heart Attack Risk from Voice, Here’s the Explanation

Suara.com – A heart attack occurs when one of the arteries leading to the heart becomes blocked, causing blood flow to suddenly stop.

This heart attack generally triggers a squeezing sensation or sudden pain in the chest, which can radiate to the neck, jaw or back.

According to a new study, acoustic signals in patients at risk of heart attack can be picked up by artificial intelligence software years before an event.

This screening tool can identify patients who are at risk for a heart attack by analyzing the features of their voice.

Also Read:
Due to the Complications of the Corona Virus Covid-19, These Women’s Feet Must Be Amputated!

According to a recent study, this technology can capture 80 different characteristics of sound, such as rhythm, amplitude, pitch and frequency.

The researchers trialled the device using samples from 108 patients who were referred to X-rays for an assessment of the health of their coronary arteries.

Participants were asked to record three 30-second voice samples using the Vocals Health smartphone app.

Illustration of a heart attack. [Envato]
Illustration of a heart attack. [Envato]

Subjects were instructed to read the prepared text, then speak spontaneously about positive experiences for the first and second recordings.

In the final voice sample, participants were asked to speak freely about negative experiences.

Also Read:
Good News, Active Cases of Corona Virus in Pamekasan Decreased

Then, the recordings were analyzed using the Vocalis Health algorithm, an AI-based system that was trained using more than 10,000 voice recordings.

During a two-year follow-up period, 58.3 percent of participants with high voices were hospitalized for chest pain or other emergencies such as heart attacks.

In contrast, only 30.6 percent of participants with low voices returned to the hospital. Based on these data, researchers estimated that participants with high-pitched voices were 2.6 times more likely to suffer from major problems associated with coronary artery disease.

The study’s lead author, Doctor Jaskanwal Sara, from the Mayo Clinic in the United States, said telemedicine is non-invasive, cost-effective, efficient and is becoming increasingly important during a pandemic.

“We do not suggest that this sound analysis technology replaces a doctor’s diagnosis. However, we think it can be a tool for early detection of heart attacks,” said Doctor Jaskanwal Sara, quoted from Express.

Using this technology, the research team was able to identify six features associated with coronary artery disease, which were combined into a single score to be expressed as a number between minus one and one.

“We ourselves cannot hear these particular features. This technology uses machine learning to measure something that is not easily measured using the human brain and human ear,” he explained.