A team of USC researchers is developing an AI-driven platform that aims to predict both risk factors and protective factors for depression and suicide using patterns in brain activity and behavior. The project, supported by a contract award of up to $4 million from the Advanced Research Projects Agency for Health (ARPA-H), will continuously monitor college student mental health through a combination of lab tests, wearable sensors and smartphone data. Its findings could help pinpoint when a mental health crisis is imminent and how best to intervene. The new study, known as Scalable Evaluation of Neurobehavioral Trajectories in Everyday Life (SENTINEL), will track approximately 210 USC students for up to 36 months. Study participants will be evenly split between three groups: students with no mental health concerns, students with depression, and students with depression that includes suicidal thoughts or behaviors. Once enrolled in SENTINEL, participants will complete a series of tasks and tests in the laboratory every six months. These include EEG (a measure of brain activity), eye tracking and detailed psychiatric questionnaires. They will also be given wearable devices, such as an Oura Ring or Fitbit, to gather additional data on their daily habits and well-being. These devices provide information on heart rate, heart rate variability, skin temperature, skin conductance (a measure of stress), blood oxygen levels, physical activity, and sleep stages and duration. In addition, students will complete weekly questionnaires through a smartphone app about their stress, anxiety, mood, overall well-being, and daily challenges and positive experiences. They will answer some questions by voice so researchers can analyze speech patterns. With consent, researchers will also collect smartphone data on app usage, communication patterns and attention. The study is designed to capture not just periods of distress, but also moments when students are doing well. Researchers aim to rapidly translate its findings into tools that can help clinicians identify risk and intervene early.