Anthony Law, MD, PhD, an assistant professor in the department of otolaryngology at Atlanta-based Emory University School of Medicine, has developed an AI model to detect throat cancer in a patient’s voice.
The AI model, a deep neural network, was trained on a database of 15,000 voice recordings and is “about 93% successful” at identifying patients who have a mass in their larynx, according to a June 16 news release from the university.
“It’s a very easy sign for trained laryngologists to hear that someone has a mass in their larynx,” Dr. Law said in the release. “But [primary care doctors] don’t have that extra training to hear concerning voice versus non-con concerning voice and screen out people who have voice changes due to a cold versus voice changes due to cancer.”
The model was trained across demographics to ensure diagnoses remain “fair and equitable,” Dr. Law said. The technology has been designed for in-clinic use through an app and takes about five minutes to record 10 voice prompts.
[…] Separate from the Frontiers study, an Emory University laryngologist built an in-clinic app that records ten short prompts and flags whether a patient likely has a laryngeal mass—a proxy for underlying cancer—reporting about 93% accuracy on internal testing so far. The system was trained on roughly 15,000 voice samples across demographics. Emory News; Becker’s. […]
[…] Separate from the Frontiers study, an Emory University laryngologist built an in-clinic app that records ten short prompts and flags whether a patient likely has a laryngeal mass—a proxy for underlying cancer—reporting about 93% accuracy on internal testing so far. The system was trained on roughly 15,000 voice samples across demographics. Emory News; Becker’s. […]