Voice, a Biomarker to Detect Diseases

To the general public, it may sound like something out of a science fiction movie: diagnosing serious diseases such as cancer by listening to someone’s voice. But in fact, researchers funded by the National Institutes of Health (NIH) are now investigating whether changes in a person’s voice could serve as a new biomarker in clinical care for detecting illnesses early.

The project, called Voice as a Biomarker of Health, will include 12 research institutions and is being funded by the Bridge to Artificial Intelligence (Bridge2AI) program out of the NIH Common Fund. The project will use machine learning to build a database of vocal biomarkers, and then use the science of acoustic analysis to identify changes—such as pitch, amplitude, cadence, and words per minute—that could become a low-cost diagnostic tool, alongside other clinical tests.

“Voice has the potential to be a biomarker for several health conditions,” said Yael Bensoussan, MD, Assistant Professor in the Department of Otolaryngology and Director of the University of South Florida (USF) Health Voice Center at the USF Morsani College of Medicine and co-principal investigator on the Voice as a Biomarker of Health project. One of those conditions is laryngeal cancer.

The goal of NIH‘s Bridge2AI program is to foster advances in biomedical research that set the stage for widespread adoption of AI. AI in early detection includes the ability of a computer to perform tasks commonly associated with human intelligence, and using programmed data to make predictions or decisions. A key step is generating new datasets and best practices for machine learning analysis. AI “has a lot of promise, but it’s very limited now,” said Grace C.Y. Peng, PhD, Co-Coordinator of Bridge2AI and Director of Mathematical Modeling, Simulation, and Analysis at the National Institute of Biomedical Imaging and Bioengineering, a component of the NIH.

Source: https://journals.lww.com/