A research team at the University of Southern California (USC) is developing an AI-powered platform aimed at predicting risk factors and protective factors related to depression and suicide. This initiative is funded by a grant of up to $4 million from the Advanced Research Projects Agency for Health (ARPA-H).

Read More

The project, called Scalable Evaluation of Neurobehavioral Trajectories in Everyday Life (SENTINEL), will track approximately 210 USC students over a period of 36 months. The participants will include three groups: students with no mental health concerns, those experiencing depression, and those with depression accompanied by suicidal thoughts or behaviors. This dual focus on risk and resilience aims to provide a fuller understanding of mental health challenges in college students.

Data collection will involve monitoring brain activity, sleep patterns, smartphone usage, and other factors. Participants will undergo lab tests every six months, including EEG assessments and eye tracking, and utilize wearable devices to monitor daily health metrics such as heart rate and sleep duration. Additionally, weekly questionnaires completed through a smartphone app will assess stress, anxiety, mood, and overall well-being.

Principal investigator Shrikanth Narayanan, a professor at USC, emphasized the project's goal of addressing significant mental health questions through interdisciplinary collaboration. Assal Habibi, an associate professor of psychology involved in the study, noted that understanding early warning signs of mental health issues among college students is crucial.

The study's observational nature is intended to translate findings into clinical tools for identifying those at risk, aiding timely interventions. Researchers aim to identify patterns that signal increased vulnerability to depression, which could lead to tailored therapeutic options.

This research is aligned with ARPA-H’s EVIDENT initiative, focusing on the use of real-time clinical data to enhance behavioral health understanding. The findings may extend beyond college students to support other high-stress populations, including healthcare workers and first responders.