USF hosts the first summer course on the use of language and AI as a diagnostic toolArticle title

From Sarah Smith, University marketing and communication

Recent developments in artificial intelligence are allowing researchers to assess how the voice can help predict and detect disease early. In collaboration with the USF AI+X Institute, USF Health is supporting the training of the next generation of healthcare professionals by launching the first course to explore how the voice can be used as a biomarker of health – helping to identify neurological, respiratory and psychiatric conditions.

“While there are many programs that teach students how to use imaging and other clinical data, the ability to understand and analyze speech data puts these students in a unique position,” says laryngologist Yael Bensoussan, MD, assistant professor of otolaryngology at USF Health. “There are many startups in many different industries looking for this expertise because it’s still very rare to find. We’re positioning them to be ready to apply for these jobs.” Bensoussan is co-principal investigator of the National Institutes of Health-funded Bridge2AI-Voice program, a consortium of 50 professionals including clinicians, speech therapists, computer scientists, AI scientists and bioethicists working on standardizing AI datasets and their ethical use in healthcare.

As part of the program, USF is one of four universities across the country selected to host the first Voice AI Summer School. The five-week course trains students from undergraduate to medical school to process speech audio, recognize acoustic and linguistic features such as hoarseness or tremors, and train and test algorithms that can help with diagnosis.

Trevor Pharr, a rising sophomore at the USF Health Morsani College of Medicine, says his love of singing led to his interest in becoming a doctor specializing in otolaryngology. Pharr double majored in biochemistry and music at the University of North Carolina, Chapel Hill, and is now studying the groundbreaking field of speech AI.

“I never thought I would study this, but it’s a happy coincidence that this really, really good voice study is happening now because it’s something I’ve been interested in for years,” Pharr said. “I’m just happy to be part of this experience.”

Forty-five students from across the country have been selected for the intensive program, 15 of whom are students at USF. They will be divided into teams from engineering and health fields. Students will practice collecting voice data and are invited to voice their own voices, which Bensoussan said will deepen their understanding of the technology even further.

“Sometimes engineers are presented with data without necessarily understanding how it was collected,” Bensoussan said. “I think it’s really important to understand how it was collected, and the best way to understand that is to do it yourself.”

Faculty members of USF’s AI+X Institute – a multidisciplinary research and development center focused on AI – teach courses on the intricacies of developing AI models using acoustic data for healthcare purposes. Many of them come from USF’s Department of Computer Science and Engineering, including Distinguished University Professor and Department Chair Sudeep Sarkar. He is also the Associate Vice President for Research and Innovation at USF and serves as the Director of the AI+X Institute.

“Our engagement underscores our commitment to nurturing the next generation of AI researchers and practitioners from diverse disciplines – the ‘+X’ part of our identity,” said Sarkar. “This program provides participants with a unique platform to gain hands-on experience, engage with cutting-edge research, and collaborate with leading experts in the unique problem space of language-based medical diagnosis.”

Shriamana Ganesh-Sudhadhakai is a second-year bioinformatics student at Georgia Tech University. He is attending the Voice AI Summer School in hopes of entering the bioinformatics industry.

“Bioinformatics is somewhere between biology and computer science,” said Ganesh-Sudhadhakai. “I want to be in the industry and focus on research and product development, so for me the ideal goal is to learn all the technologies currently available.”

Bensoussan hopes the benefits of the Voice AI Summer School will extend far beyond the enrolled students.

“A lot of our project is open science, so we hope that the curriculum we develop will be of use to a wider community because that’s how things get discovered.”