Alina Hvostikova/Stocksy
- A new study shows that voice analysis, including characteristics such as pitch and amplitude, can predict the risk of coronary heart disease and its complications.
- This novel approach involves the use of artificial intelligence algorithms to analyze voice samples collected with the help of a mobile application.
- This voice analysis technology could serve as a cost-effective and convenient approach for screening individuals at risk of coronary artery disease.
The early detection of coronary artery disease using this voice analysis approach could potentially improve patient outcomes. This approach relies on the collection of voice samples using a mobile application and could serve as a cost-effective and noninvasive method to remotely screen individuals at risk of coronary artery disease.
Voice samples analysis could be used as a preliminary tool for identifying patients in need of closer attention for coronary artery disease events.
The study’s co-author, Dr. Jaskanwal Deep Singh Sara, a research fellow at Mayo Clinic, says, “We’re not suggesting that voice analysis technology would replace doctors or replace existing methods of healthcare 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.”
The study was presented at the American College of Cardiology conference that took place in April 2-4, 2022, in Washington, DC, and was simultaneously published in the Mayo Clinic Proceedings.
Coronary artery disease is characterized by the buildup of plaque in the inner walls of coronary arteries, which supply blood to the heart. The buildup of plaque leads to the narrowing or blocking of these arteries, reducing blood supply to the heart. Complications associated with coronary artery disease
The authors of the present study had
In the current study, the researchers wanted to determine if these previously characterized voice features could predict the risk of coronary artery disease.
The present study included 108 individuals referred for coronary angiography, which involves the use of X-rays to assess the condition of coronary arteries.
The researchers obtained voice recordings from each person with the help of a mobile application. The patients were asked to provide three different vocal samples — reading a text, describing a negative experience, and describing a positive experience.
The researchers used artificial intelligence to analyze these audio samples and extract voice features that the researchers had found to be associated with coronary artery disease in their previous study. They derived a single biomarker score for each participant using these voice features. The participants were then categorized into two groups based on having either a low or high biomarker score.
After a follow-up period of 2 years, the researchers found that people with a higher biomarker score were at 2.6 times higher risk of experiencing or being hospitalized for symptoms associated with coronary artery diseases, such as chest pain or a heart attack.
Moreover, individuals with higher biomarker scores were 3.1 higher risk to be diagnosed with coronary artery disease during a subsequent angiography or have a positive exercise stress test, an indicator of coronary artery disease risk.
Voice analysis has significant promise in the burgeoning field of telemedicine. Dr. Alan Sugrue, a researcher at Mayo Clinic, wrote in an editorial, “The future of voice analysis and human health is teeming with countless opportunities. Because all telecommunications today are generally digital, voice analysis could be easily integrated into current technological platforms (such as a smartphone) with either analysis by software on the platform or transmission of digital voice recordings to a central procession area.”
“Artificial intelligence has the potential to learn your voice and its myriad variations and thereby determine whether substantial and subtle changes may correlate with disease that is either subacute or acute.”
Similarly, Dr. Sara told MNT, “The evolution of digital biomarkers and the delivery of healthcare remotely from medical facilities has accelerated during the current [COVID-19] global pandemic. The coupling of voice signal analysis with artificial intelligence and machine-based learning offers an exciting and innovative solution to the growing demand for telemedicine.”
“Voice biomarkers, derived from the analysis and extraction of characteristic acoustic and linguistic vocal patterns, have been shown to be associated with cardiovascular, neurologic, and psychiatric diseases and even [COVID-19] itself.”
– Dr. Sara
The mechanism explaining the ability of these voice features to serve as a biomarker for coronary artery disease remains unknown.
Voice production and its control are also to a large extent unconscious and modulated by the autonomic nervous system. Thus, the autonomic nervous system, being a common denominator underlying vocal control and heart function, could explain the ability of voice features to serve as a proxy for cardiovascular health.
The
The researchers note that the artificial intelligence platform used in the study was initially trained using voice samples collected from individuals in Israel.
Although the present study was conducted using voice samples from English-speaking individuals residing in the Midwestern United States, the scientist cautioned that more research is necessary to determine if these findings can be generalized to larger and more diverse demographics who speak different languages.
Dr. Sara says, “It’s definitely an exciting field, but there’s still a lot of work to be done. We have to know the limitations of the data we have, and we need to conduct more studies in more diverse populations, larger trials, and more prospective studies like this one.”
As with other telehealth technologies, security and privacy challenges associated with this new approach also need to be addressed before its widespread adoption.