MULTICAST

The main goal of this study is to improve our understanding of who might face mental health problems or need further psychiatric help after leaving a psychiatric hospital.

The overall goal of this study is to improve the ability to predict mental health after discharge from a psychiatric unit. The motivation is to improve the prediction of who is at risk for psychiatric disorders, suicidal ideation, or hospital readmission after discharge from an inpatient hospitalization.

We use artificial intelligence (AI) to identify signs of psychiatric symptoms from video and audio data. Physical signs such as tone of voice, head movements, eye movements, and facial expressions provide information about mental health. In addition, we plan to combine the information extracted from the video and audio data with clinical and behavioral data collected during the baseline assessment and with the CORA app to further improve the accuracy of the prediction. We believe our model will enable more accurate risk prediction and help clinicians identify at-risk patients so they can initiate the right treatment when patients need it most, reducing patient burden and downstream healthcare costs.

For more information about this study:
Kacey Ferguson, Senior Research Coordinator, Program Coordinator
Kacey.Ferguson@nyulangone.org

Or the MULTICAST Team at:
MULTICAST@nyulangone.org