DENVER — A new research study led by Johnny Kim, a professor at the University of Denver’s Graduate School of Social Work, is looking to determine whether artificial intelligence (AI) can be used to screen for anxiety and depression in youth.
The study is testing software from Ellipsis Health, which recruited help from Kim. The vocal biomarker technology, which has already been validated for adults, allows students to engage in a 90-second conversation about their feelings on any device with a microphone and an internet connection. The recordings are analyzed by the software’s algorithm to flag students for anxiety and depression.
“It looks at both the acoustics in terms of how words are said, as well as the semantics, what is said specifically,” explained Kim. "It does not rely on a trained clinician to be able to make observations through non-verbal cues, or rely on the adolescent to be able to truthfully answer the questions in order to be able to screen appropriately and accurately.”
The mental health evaluation process for children has become even more important in Colorado. Governor Jared Polis signed the Sixth Through Twelfth Grade Mental Health Screening Act in 2023, which mandates mental health evaluations for each Colorado student in those grades.
Kim is hoping to collaborate with the Colorado Office of Behavioral Health, which oversees the new evaluations, to streamline the process.
“It's very challenging, from a time perspective, to be able to screen thousands of students,” said Kim. “If this is shown to be just as effective as the current approaches, or maybe even more effective, the implications can be really incredible for school districts that serve thousands of students."
The group is hoping to wrap up data collection this month and start analyzing the data in January, before being able to publish their findings in academic journals.
Denver7 is committed to making a difference in our community by standing up for what's right, listening, lending a helping hand and following through on promises. See that work in action, in the videos above.