University of South Florida Health is leading the way on a project that could help diagnose cancer and depression through the sound of a patient’s voice.
USF Health is partnering with New York City-based Weill Cornell Medicine, along with 10 other institutions across North America, on the project. The entities are also working with New York-based biotech startup Owkin.
“The potential for using voice and sounds together with advanced AI algorithms to accurately diagnose certain diseases is incredible,” said Dr. Olivier Elemento, professor of physiology and biophysics at Weill Cornell Medicine, in a statement. “Our future findings could lead to a revolution in health care.”
The project is through the National Institutes of Health and could span up to four years. The NIH will deploy $3.8 million in the project’s first year and give up to $14 million over the following three years, contingent upon NIH appropriations by Congress.
The project, called “Voice as a Biomarker of Health,” is part of the NIH’s program focused on using AI to tackle biomedical problems. A database of voices will be ethically sourced and protected with patient privacy, according to officials. Machine learning models will be trained to spot diseases by detecting changes in the patients’ voices. It would be used in addition to other traditional clinical methods.
The project will focus on five potential categories:
- Voice disorders such as laryngeal cancers, vocal fold paralysis and benign laryngeal lesions
- Neurological and neurodegenerative disorders, such as Alzheimer’s, Parkinson’s, stroke and ALS
- Mood and psychiatric disorders, such as depression, schizophrenia, bipolar disorders
- Respiratory disorders, such as pneumonia and COPD
- Pediatric voice and speech disorders, such as speech and language delays and autism
“Voice has the potential to be a biomarker for several health conditions,” said Dr. Yaël Bensoussan, director of USF Health Voice Center, in a statement. “Creating an effective framework that incorporates huge datasets using the best of today’s technology in a collaborative manner will revolutionize the way that voice is used as a tool for helping clinicians diagnose diseases and disorders.”
USF has received many NIH grants over the last several years, which do more than elevate the university on a national scale. For every roughly $1 of funding, $2.60 of economic activity occurred. In 2021, USF garnered more than $92 million across 135 grants.
