CCMapper

An adaptive NLP-based free-text chief complaint mapping algorithm

Mohammad Samie Tootooni, Kalyan S Pasupathy, Heather A. Heaton, Casey M. Clements, Mustafa Sir

Research output: Contribution to journalArticle

Abstract

Objective: Chief complaint (CC) is among the earliest health information recorded at the beginning of a patient's visit to an emergency department (ED). We propose a heuristic methodology for automatically mapping the free-text data into a structured list of CCs. Methods: A comprehensive structured list categorizing CCs was developed by experienced Emergency Medicine (EM) physicians. Using this list, we developed a natural language processing-based algorithm, referred to as Chief Complaint Mapper (CCMapper), for automatically mapping a CC into the most appropriate category (ies). We trained and validated CCMapper using free-text CC data from the Mayo Clinic ED in Rochester, MN. We developed a consensus-based validation approach to handle both indifferences and disagreements between the two EM physicians who manually mapped a random sample of free-text CCs into categories within the structured list. Results: The kappa statistic demonstrated a high level of agreement (κ = 0.958) between the two physicians with less than 2% human error. CCMapper achieved a total sensitivity of 94.2% with a specificity of 99.8% and F-score of 94.7% on the validation set. The sensitivity of CCMapper when mapping free-text data with multiple CCs was 82.3% with a specificity of 99.1% and total F-score of 82.3%. Conclusion: Due to its simplicity, high performance, and capability of incorporating new free-text CC data, CCMapper can be readily adopted by other EDs to support clinical decision making. CCMapper can facilitate the development of predictive models for the type and timing of important events in ED (e.g., ICU admission).

Original languageEnglish (US)
Article number103398
JournalComputers in Biology and Medicine
Volume113
DOIs
StatePublished - Oct 1 2019

Fingerprint

Hospital Emergency Service
Emergency Medicine
Physicians
Medicine
Natural Language Processing
Intensive care units
Consensus
Decision making
Health
Statistics
Processing
Heuristics
Clinical Decision-Making

Keywords

  • Emergency department
  • Free-text chief complaints
  • Heuristic
  • Human consensus-based validation
  • Iterative enhancement
  • Mapping algorithm
  • Natural language processing

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this

CCMapper : An adaptive NLP-based free-text chief complaint mapping algorithm. / Tootooni, Mohammad Samie; Pasupathy, Kalyan S; Heaton, Heather A.; Clements, Casey M.; Sir, Mustafa.

In: Computers in Biology and Medicine, Vol. 113, 103398, 01.10.2019.

Research output: Contribution to journalArticle

@article{43134e67992b4b0a9b918de1c2245aa4,
title = "CCMapper: An adaptive NLP-based free-text chief complaint mapping algorithm",
abstract = "Objective: Chief complaint (CC) is among the earliest health information recorded at the beginning of a patient's visit to an emergency department (ED). We propose a heuristic methodology for automatically mapping the free-text data into a structured list of CCs. Methods: A comprehensive structured list categorizing CCs was developed by experienced Emergency Medicine (EM) physicians. Using this list, we developed a natural language processing-based algorithm, referred to as Chief Complaint Mapper (CCMapper), for automatically mapping a CC into the most appropriate category (ies). We trained and validated CCMapper using free-text CC data from the Mayo Clinic ED in Rochester, MN. We developed a consensus-based validation approach to handle both indifferences and disagreements between the two EM physicians who manually mapped a random sample of free-text CCs into categories within the structured list. Results: The kappa statistic demonstrated a high level of agreement (κ = 0.958) between the two physicians with less than 2{\%} human error. CCMapper achieved a total sensitivity of 94.2{\%} with a specificity of 99.8{\%} and F-score of 94.7{\%} on the validation set. The sensitivity of CCMapper when mapping free-text data with multiple CCs was 82.3{\%} with a specificity of 99.1{\%} and total F-score of 82.3{\%}. Conclusion: Due to its simplicity, high performance, and capability of incorporating new free-text CC data, CCMapper can be readily adopted by other EDs to support clinical decision making. CCMapper can facilitate the development of predictive models for the type and timing of important events in ED (e.g., ICU admission).",
keywords = "Emergency department, Free-text chief complaints, Heuristic, Human consensus-based validation, Iterative enhancement, Mapping algorithm, Natural language processing",
author = "Tootooni, {Mohammad Samie} and Pasupathy, {Kalyan S} and Heaton, {Heather A.} and Clements, {Casey M.} and Mustafa Sir",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.compbiomed.2019.103398",
language = "English (US)",
volume = "113",
journal = "Computers in Biology and Medicine",
issn = "0010-4825",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - CCMapper

T2 - An adaptive NLP-based free-text chief complaint mapping algorithm

AU - Tootooni, Mohammad Samie

AU - Pasupathy, Kalyan S

AU - Heaton, Heather A.

AU - Clements, Casey M.

AU - Sir, Mustafa

PY - 2019/10/1

Y1 - 2019/10/1

N2 - Objective: Chief complaint (CC) is among the earliest health information recorded at the beginning of a patient's visit to an emergency department (ED). We propose a heuristic methodology for automatically mapping the free-text data into a structured list of CCs. Methods: A comprehensive structured list categorizing CCs was developed by experienced Emergency Medicine (EM) physicians. Using this list, we developed a natural language processing-based algorithm, referred to as Chief Complaint Mapper (CCMapper), for automatically mapping a CC into the most appropriate category (ies). We trained and validated CCMapper using free-text CC data from the Mayo Clinic ED in Rochester, MN. We developed a consensus-based validation approach to handle both indifferences and disagreements between the two EM physicians who manually mapped a random sample of free-text CCs into categories within the structured list. Results: The kappa statistic demonstrated a high level of agreement (κ = 0.958) between the two physicians with less than 2% human error. CCMapper achieved a total sensitivity of 94.2% with a specificity of 99.8% and F-score of 94.7% on the validation set. The sensitivity of CCMapper when mapping free-text data with multiple CCs was 82.3% with a specificity of 99.1% and total F-score of 82.3%. Conclusion: Due to its simplicity, high performance, and capability of incorporating new free-text CC data, CCMapper can be readily adopted by other EDs to support clinical decision making. CCMapper can facilitate the development of predictive models for the type and timing of important events in ED (e.g., ICU admission).

AB - Objective: Chief complaint (CC) is among the earliest health information recorded at the beginning of a patient's visit to an emergency department (ED). We propose a heuristic methodology for automatically mapping the free-text data into a structured list of CCs. Methods: A comprehensive structured list categorizing CCs was developed by experienced Emergency Medicine (EM) physicians. Using this list, we developed a natural language processing-based algorithm, referred to as Chief Complaint Mapper (CCMapper), for automatically mapping a CC into the most appropriate category (ies). We trained and validated CCMapper using free-text CC data from the Mayo Clinic ED in Rochester, MN. We developed a consensus-based validation approach to handle both indifferences and disagreements between the two EM physicians who manually mapped a random sample of free-text CCs into categories within the structured list. Results: The kappa statistic demonstrated a high level of agreement (κ = 0.958) between the two physicians with less than 2% human error. CCMapper achieved a total sensitivity of 94.2% with a specificity of 99.8% and F-score of 94.7% on the validation set. The sensitivity of CCMapper when mapping free-text data with multiple CCs was 82.3% with a specificity of 99.1% and total F-score of 82.3%. Conclusion: Due to its simplicity, high performance, and capability of incorporating new free-text CC data, CCMapper can be readily adopted by other EDs to support clinical decision making. CCMapper can facilitate the development of predictive models for the type and timing of important events in ED (e.g., ICU admission).

KW - Emergency department

KW - Free-text chief complaints

KW - Heuristic

KW - Human consensus-based validation

KW - Iterative enhancement

KW - Mapping algorithm

KW - Natural language processing

UR - http://www.scopus.com/inward/record.url?scp=85070968393&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070968393&partnerID=8YFLogxK

U2 - 10.1016/j.compbiomed.2019.103398

DO - 10.1016/j.compbiomed.2019.103398

M3 - Article

VL - 113

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

SN - 0010-4825

M1 - 103398

ER -