Extraction of structured data from unstructured palliative care consult questions using the MedSpacy NLP Library

Poster Number: P100
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Natural Language Processing, Information Extraction, Internal Medicine or Medical Subspecialty
Primary Track: Applications
Programmatic Theme: Clinical Research Informatics

Accessing detailed structured data from electronic health records (EHRs) for research and quality improvement is a significant challenge for smaller departments in less resourced institutions. As such, extracting maximal information from available data is key. Using Medspacy, an open source clinical natural language processing library for Python, we developed a rule-based algorithm to extract structured patient data from unstructured palliative care consult questions to create more granular departmental data.

Speaker(s):
Kent McCann, MD
Baystate Medical Center