S108: Panel - Clinical Text Mining in Mental Health

10:30 AM–12:00 PM Nov 20, 2019 (US - Eastern)

Georgetown East


It has been shown that application of natural language processing (NLP) and text mining of clinical notes in electronic health records (EHRs) can help facilitate deeper and more accurate phenotyping in many disease areas. These methods are particularly useful in the context of mental illness, where disorders are highly heterogeneous and lacking diagnostic lab tests, medications are non-specific, and diagnosis codes are used inconsistently by different care providers, if not left out altogether. However, NLP approaches in mental health EHRs also pose a number of challenges, some common across all of medicine, some unique to mental illness.
The assembled panelists represent a geographically and scientifically diverse set of perspectives in the field of clinical NLP for mental health. They will discuss how their respective groups have addressed both scientific and infrastructural challenges and how they have helped to advance the field. They will also describe lessons learned as well as tools and artifacts they have developed that will be of use to other groups who are interested in similar questions.

Learning Objective: Understand the challenges and opportunities in using natural language processing of clinical text in mental health records.
Identify cutting edge methods and community resources in this space.


Jessica Tenenbaum (Presenter)

Ramakanth Kavuluru, University of Kentucky
Thomas McCoy, Massachusetts General Hospital
Ozlem Uzuner, George Mason University
Sumithra Velupillai, King's College

Presentation Materials: