Stony Brook University study uses AI to track public mental health via social media

At Stony Brook University on Long Island, researchers are using artificial intelligence to measure a person’s well-being based on social media activity.

A study led by Stony Brook researchers recently published in Nature Digital Medicine found the accuracy of the AI assessment was comparable to traditional polling. It also sampled more people in a shorter time and generated more precise information at a lower cost.

When comparing pre and post-COVID-19, findings show a shift in anxiety and depression on Long Island. The same held true nationally in response to other major events, including the murder of George Floyd.

The team is waiting on data from this year to analyze the current anxiety and depression levels associated with local and world events like the earthquake and the Israel-Hamas war.

"This is really important because recently there's been a rise in causes of death related to mental health conditions. Suicide mortality has been on the rise, especially among adolescents and youth. Opioid mortality as well, and these are conditions that you can't understand how they change just by monitoring biological data from hospitals," said Associate Professor Andrew Schwartz.

Researchers also hope the tool can also help mental health professionals better understand the needs of their community.