Description
There is a wealth of information that is documented in the electronic health record, however, they are not discretely captured. The abundance of detailed information is stored in narrative text that aids clinician in better understanding disease manifestation and stage of the disease. Costs of extracting symptoms manually can be offset by utilizing natural language processing tools. We can use these extractions to build a robust set of features for a clinical prediction algorithm.
Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Understand the feasibility of utilizing NLP tools for feature / symptom extraction from clinical narratives and its implications.
Authors:
Myung Woo (Presenter)
Duke
Dylan Liu, Duke
Jessica Tenenbaum, Duke
Presentation Materials: