Predicting Clinical Mental Health Care Utilization Outcomes using Structured and Textual Data from Electronic Health Records
This study aims to predict three important outcomes related to mental health care: (1) how long someone might stay in a mental health hospital, (2) whether a person will need to return to the hospital after being discharged, and (3) the costs after someone is first referred. We are combining different types of data, including both structured information and full-text records, using modern artificial intelligence techniques.