Natural Language Processing for Clinical Applications: Challenges and Opportunities, Carol Friedman, PhD, Professor of Biomedical Informatics, Columbia University

Dr. Friedman is an internationally recognized expert on the use of advanced natural language processing techniques to extract, structure and encode information in clinical documents. She has developed the MedLEE system, which is used in a variety of medical domains to extract clinical, biomolecular and genetic information from biomedical documents. She is a Fellow of the American College of Medical Informatics.

Abstract: Clinical data is critical to patient care, yet most patient information is in the form of text and, therefore, cannot be utilized by automated processes. Natural language processing (NLP) is a method that transforms text into structured encoded data, which is a form that can be reliably accessed by clinical applications needed to improve the health care process. This talk will discuss opportunities and challenges for NLP in the clinical domain. In addition, we will present an overview of the MedLEE NLP system that we developed as well as demonstrate automated clinical applications based on use of MedLEE.