New study uses AI to improve treatment of teens for depression

Children’s Mercy is recruiting volunteers for the 3-month study.

Anxiety and depression are becoming increasing concerns among the youth. A popular treatment path is to prescribe anti-depressant medicines, but it often takes weeks to know if the treatment is even effective.

“Many patients and their families often suffer through weeks of trial and error,” according to a blog from Children’s Mercy Research Institute (CMRI) in Kansas City, Missouri.

To that end, CMRI is conducting a research study with the Mayo Clinic to find effective treatments for patients in the shortest time possible using an AI/ML model.

The study, GOLDILOKs PRISM, seeks to determine how accurately an AI/ML model can predict a patient’s response to fluoxetine, commonly known as Prozac, after a few weeks of treatment. It does so by comparing the teen’s first three months of Prozac treatment with the outcome predicted by the AI/ML model.

The study currently is recruiting youth from ages 12 to 18 that have been diagnosed with depression and been prescribed Prozac.

“There is no one-size-fits-all treatment for depression,” said Dr. Catherine Koertje, who leads the study at Children’s Mercy, in the blog. “The end product of this research initiative is to develop a tool that clinicians, patients, and their families can use to make more informed, individualized, and timely decisions about how to best treat a teen’s depression.”

This study is different in that it applies research into the clinical setting to shorten the time frame than going through a typical clinical trial, Koertje said.

The study also seeks to identify factors that can affect Prozac’s efficacy. These factors include genetic, biochemical, proteomic, lifestyle and other data to generate “signatures” linked to clinical outcomes.

Patients’ heart rate, sleep and physical activity will be monitored, which researchers will score and interpret. The data is downloaded and stored for customized reports for daily evaluations.

Future plans include looking at known genes and find unknown genes affected by selective serotonin reuptake inhibitors like Prozac.