Leveraging Informatics to Improve Mental Health Services
This project aims to (1) develop innovative data science methods and tools tailored for data generated in mental health services; (2) enhance prevention and intervention of mental health problems for vulnerable population.
Funding: Startup Grant
Understanding Patient Safety Events via Multi-label Text Classification and Semantic Representation
Medical errors are the third leading causes of death in the US. One of the key steps to improve patient safety is through patient safety event reporting. A patient safety event often captures a number of critical factors that are involved during the care when incidents, near miss, and unsafe conditions occured. Clinicians and risk management team can generate actionable knowledge by harnessing useful information from the reports, aka., Root Cause Analysis (RCA).
We found that the narrative data of reports (e.g., event description in free text) carries valuable information (e.g., interconnected procedures, shared responsibilities, contributing factors, etc.) that may disclose vulnerabilities of the health system. However, the traditional RCA is time consuming. Inspired by psycholinguistic studies, semantic web ontology, and machine learning, we developed methods to facilitate the RCA.
Funding: NIH/AHRQ and the University of Texas System (PI: Yang Gong, MD, PhD)