Artificial intelligence may soon help doctors diagnose and treat diseases, including cancer and depression. All based on the sound of a patient’s voice. Twelve leading research institutions, including the University of South Florida, are launching a landmark National Institutes of Health-funded academic project. This project may end up establishing voice as a biomarker used in clinical care.
The University of South Florida is the lead institution on the project in collaboration with Weill Cornell Medicine in NYC. There are also 10 other institutions in the United States and Canada as well as French-American AI biotech startup Owkin. The first year of the project includes $3.8 million from the NIH. There is subsequent funding over the following three years contingent upon annual NIH appropriations by Congress. This could bring the overall award to $14 million.
How to detect diseases using voice
Called Voice as a Biomarker of Health, the project is one of several recently funded by the NIH Common Fund’s Bridge2AI program. This program is designed to use AI to tackle complex biomedical challenges. The voice project will aim to build an ethically sourced database of diverse human voices while protecting patient privacy. Using this data, machine learning models will be trained to spot diseases by detecting changes in the human voice. This could, in turn, empower doctors with a low-cost diagnostic tool they can use with other clinical methods.
Based on the existing literature, the research team has identified five disease categories they are going to focus on:
- Voice disorders (laryngeal cancers, vocal fold paralysis, benign laryngeal lesions)
- Neurological and neurodegenerative disorders (Alzheimer’s, Parkinson’s, stroke, ALS)
- Mood and psychiatric disorders (depression, schizophrenia, bipolar disorders)
- Respiratory disorders (pneumonia, COPD)
- Pediatric voice and speech disorders (speech and language delays, autism)
Related: Schools to Adopt Tom Brady Inspired Health and Wellness Curriculum
Preliminary work with voice data has been promising. However, there are limitations to integrating voice as a biomarker in clinical practice due to small datasets. This is due to the ethical concerns around data ownership and privacy, bias and lack of diversity of the data. To solve these, the Voice as a Biomarker of Health project is creating a large, high-quality, multi-institutional and diverse voice database that is linked to identity-protected/unidentifiable biomarkers from other data, such as demographics, medical imaging and genomics.
Project leads
Supported by AI experts, bioethicists and social scientists, the project aims to transform our fundamental understanding of diseases. It could introduce a revolutionary new method of diagnosing and treating diseases into clinical settings. The human voice is low-cost, easy to store and readily available. So diagnosing diseases through the voice using AI could prove a transformative step in precision medicine and accessibility.
Dr Yaël Bensoussan, MD, from USF Health Morsani College of Medicine and Olivier Elemento, PhD from Weill Cornell Medicine are co-principal investigators for the project.
“Voice has the potential to be a biomarker for several health conditions,” said Dr. Bensoussan, assistant professor in the Department of Otolaryngology and director of USF Health Voice Center at the USF Health Morsani College of Medicine. “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.”
Click here to read more and see all lead investigators on the project.
