Title Information
Title
Automated extraction of substance use information from clinical texts
Name: Personal
Name Part
Wang, Yan
Role
Role Term: Text (marcrelator)
author
affiliation
University of Minnesota. Institute for Health Informatics
Name: Personal
Name Part
Chen, Elizabeth S.
Role
Role Term: Text (marcrelator)
author
affiliation
Brown University. Center for Biomedical Informatics
Name: Personal
Name Part
Arsoniadis, Elliot
Role
Role Term: Text (marcrelator)
author
affiliation
University of Minnesota. Institute for Health Informatics
Name: Personal
Name Part
Carter, Elizabeth W.
Role
Role Term: Text (marcrelator)
author
affiliation
University of Vermont. Center for Clinical and Translational Science
Name: Personal
Name Part
Lindemann, Elizabeth
Role
Role Term: Text (marcrelator)
author
affiliation
University of Minnesota. Institute for Health Informatics
Name: Personal
Name Part
Sarkar, Indra Neil
Role
Role Term: Text (marcrelator)
author
affiliation
Brown University. Center for Biomedical Informatics
Name: Personal
Name Part
Melton, Genevieve B.
Role
Role Term: Text (marcrelator)
author
affiliation
University of Minnesota. Institute for Health Informatics
Type of Resource
text
Genre (aat)
papers (documents)
Origin Information
Date Created (keyDate="yes", encoding="w3cdtf")
2015
Physical Description
Extent
p. 2121-2130
Language
Language Term: Code (ISO639-2B)
eng
Abstract
Within clinical discourse, social history (SH) includes important information about substance use (alcohol, drug, and nicotine use) as key risk factors for disease, disability, and mortality. In this study, we developed and evaluated a natural language processing (NLP) system for automated detection of substance use statements and extraction of substance use attributes (e.g. temporal and status) based on Stanford Typed Dependencies. The developed NLP system leveraged linguistic resources and domain knowledge from a multi-site social history study, Propbank and the MiPACQ corpus. The system attained F-scores of 89.8, 84.6 and 89.4 respectively for alcohol, drug, and nicotine use statement detection, as well as average F-scores of 82.1, 90.3, 80.8, 88.7 96.6, and 74.5 respectively for extraction of attributes. Our results suggest that NLP systems can achieve good performance when augmented with linguistic resources and somain knowledge when applied to a wide breadth of substance use free text clinical notes.
Note
This work is funded by National Library of Medicine/National Institutes of Health grant R01LM011364.
Subject (LCSH)
Topic
Drug use
Subject (LCSH)
Topic
Alcohol use
Subject (LCSH)
Topic
Tobacco use
Related Item: Host (displayLabel="Published in")
Title Information
Title
AMIA Annual Symposium Proceedings
Part Number
2015:2121-30
Access Condition: use and reproduction (href="https://creativecommons.org/licenses/by/4.0")
This work is licensed under a Creative Commons CC BY 4.0 License
Access Condition: logo (href="https://licensebuttons.net/l/by/4.0/88x31.png")
Identifier: PMID (displayLabel="PubMed ID")
26958312