Study: AI voice analysis can diagnose PTSD in veterans


A new study published in the journal Depression and Anxiety found that an artificial intelligence-powered voice analysis program can help diagnose post-traumatic stress disorder in military veterans.

Researchers at New York City-based NYU School of Medicine and NYU Langone Health used machine learning techniques to train the program to identify speech patterns in recorded interviews with individuals diagnosed with military service-related PTSD. Once trained, the AI was reportedly able to distinguish between the voices of those with and without PTSD with 89 percent accuracy.

Though the study did not seek to identify the underlying biological mechanisms of PTSD, the biomarkers observed in the study, including less clear speech and a lifeless tone of voice, are regularly used to help diagnose PTSD in clinical settings.

The next steps for the researchers will be to continue training and validating the AI tool using even more data, then apply for government approval for its widespread use, perhaps in the form of a PTSD-identifying smartphone app, according to Adam Brown, PhD, lead author of the study and adjunct assistant professor in the Department of Psychiatry at NYU School of Medicine.

More articles about AI:
IBM stops development, sales of Watson for Drug Discovery
Study: Crowdsourced AI faster than humans in segmenting cancerous lung tumors
Gilead, machine learning startup test AI's drug development potential

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.


Featured Whitepapers

Featured Webinars