Detection of clinically important colorectal surgical site infection using Bayesian network

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Background Despite extensive efforts to monitor and prevent surgical site infections (SSIs), real-time surveillance of clinical practice has been sparse and expensive or nonexistent. However, natural language processing (NLP) and machine learning (i.e., Bayesian network analysis) may provide the methodology necessary to approach this issue in a new way. We investigated the ability to identify SSIs after colorectal surgery (CRS) through an automated detection system using a Bayesian network. Materials and methods Patients who underwent CRS from 2010 to 2012 and were captured in our institutional American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) comprised our cohort. A Bayesian network was applied to detect SSIs using risk factors from ACS-NSQIP data and keywords extracted from clinical notes by NLP. Two surgeons provided expertise informing the Bayesian network to identify clinically meaningful SSIs (CM-SSIs) occurring within 30 d after surgery. Results We used data from 751 CRS cases experiencing 67 (8.9%) SSIs and 78 (10.4%) CM-SSIs. Our Bayesian network detected ACS-NSQIP-captured SSIs with a receiver operating characteristic area under the curve of 0.827, but this value increased to 0.892 when using surgeon-identified CM-SSIs. Conclusions A Bayesian network coupled with NLP has the potential to be used in real-time SSI surveillance. Moreover, surgeons identified CM-SSI not captured under current NSQIP definitions. Future efforts to expand CM-SSI identification may lead to improved and potentially automated approaches to survey for postoperative SSI in clinical practice.

Original languageEnglish (US)
Pages (from-to)168-173
Number of pages6
JournalJournal of Surgical Research
Volume209
DOIs
StatePublished - Mar 1 2017

Fingerprint

Surgical Wound Infection
Natural Language Processing
Colorectal Surgery
Quality Improvement
Bayes Theorem
ROC Curve
Area Under Curve
Surgeons

Keywords

  • Bayesian network
  • Colorectal surgery
  • Natural language processing
  • Surgical site infection

ASJC Scopus subject areas

  • Surgery

Cite this

@article{823294685f0e48f3b93aa3eed1f00f44,
title = "Detection of clinically important colorectal surgical site infection using Bayesian network",
abstract = "Background Despite extensive efforts to monitor and prevent surgical site infections (SSIs), real-time surveillance of clinical practice has been sparse and expensive or nonexistent. However, natural language processing (NLP) and machine learning (i.e., Bayesian network analysis) may provide the methodology necessary to approach this issue in a new way. We investigated the ability to identify SSIs after colorectal surgery (CRS) through an automated detection system using a Bayesian network. Materials and methods Patients who underwent CRS from 2010 to 2012 and were captured in our institutional American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) comprised our cohort. A Bayesian network was applied to detect SSIs using risk factors from ACS-NSQIP data and keywords extracted from clinical notes by NLP. Two surgeons provided expertise informing the Bayesian network to identify clinically meaningful SSIs (CM-SSIs) occurring within 30 d after surgery. Results We used data from 751 CRS cases experiencing 67 (8.9{\%}) SSIs and 78 (10.4{\%}) CM-SSIs. Our Bayesian network detected ACS-NSQIP-captured SSIs with a receiver operating characteristic area under the curve of 0.827, but this value increased to 0.892 when using surgeon-identified CM-SSIs. Conclusions A Bayesian network coupled with NLP has the potential to be used in real-time SSI surveillance. Moreover, surgeons identified CM-SSI not captured under current NSQIP definitions. Future efforts to expand CM-SSI identification may lead to improved and potentially automated approaches to survey for postoperative SSI in clinical practice.",
keywords = "Bayesian network, Colorectal surgery, Natural language processing, Surgical site infection",
author = "Sunghwan Sohn and David Larson and Habermann, {Elizabeth B} and Naessens, {James M} and Alabbad, {Jasim Y.} and Liu, {Hongfang D}",
year = "2017",
month = "3",
day = "1",
doi = "10.1016/j.jss.2016.09.058",
language = "English (US)",
volume = "209",
pages = "168--173",
journal = "Journal of Surgical Research",
issn = "0022-4804",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Detection of clinically important colorectal surgical site infection using Bayesian network

AU - Sohn, Sunghwan

AU - Larson, David

AU - Habermann, Elizabeth B

AU - Naessens, James M

AU - Alabbad, Jasim Y.

AU - Liu, Hongfang D

PY - 2017/3/1

Y1 - 2017/3/1

N2 - Background Despite extensive efforts to monitor and prevent surgical site infections (SSIs), real-time surveillance of clinical practice has been sparse and expensive or nonexistent. However, natural language processing (NLP) and machine learning (i.e., Bayesian network analysis) may provide the methodology necessary to approach this issue in a new way. We investigated the ability to identify SSIs after colorectal surgery (CRS) through an automated detection system using a Bayesian network. Materials and methods Patients who underwent CRS from 2010 to 2012 and were captured in our institutional American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) comprised our cohort. A Bayesian network was applied to detect SSIs using risk factors from ACS-NSQIP data and keywords extracted from clinical notes by NLP. Two surgeons provided expertise informing the Bayesian network to identify clinically meaningful SSIs (CM-SSIs) occurring within 30 d after surgery. Results We used data from 751 CRS cases experiencing 67 (8.9%) SSIs and 78 (10.4%) CM-SSIs. Our Bayesian network detected ACS-NSQIP-captured SSIs with a receiver operating characteristic area under the curve of 0.827, but this value increased to 0.892 when using surgeon-identified CM-SSIs. Conclusions A Bayesian network coupled with NLP has the potential to be used in real-time SSI surveillance. Moreover, surgeons identified CM-SSI not captured under current NSQIP definitions. Future efforts to expand CM-SSI identification may lead to improved and potentially automated approaches to survey for postoperative SSI in clinical practice.

AB - Background Despite extensive efforts to monitor and prevent surgical site infections (SSIs), real-time surveillance of clinical practice has been sparse and expensive or nonexistent. However, natural language processing (NLP) and machine learning (i.e., Bayesian network analysis) may provide the methodology necessary to approach this issue in a new way. We investigated the ability to identify SSIs after colorectal surgery (CRS) through an automated detection system using a Bayesian network. Materials and methods Patients who underwent CRS from 2010 to 2012 and were captured in our institutional American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) comprised our cohort. A Bayesian network was applied to detect SSIs using risk factors from ACS-NSQIP data and keywords extracted from clinical notes by NLP. Two surgeons provided expertise informing the Bayesian network to identify clinically meaningful SSIs (CM-SSIs) occurring within 30 d after surgery. Results We used data from 751 CRS cases experiencing 67 (8.9%) SSIs and 78 (10.4%) CM-SSIs. Our Bayesian network detected ACS-NSQIP-captured SSIs with a receiver operating characteristic area under the curve of 0.827, but this value increased to 0.892 when using surgeon-identified CM-SSIs. Conclusions A Bayesian network coupled with NLP has the potential to be used in real-time SSI surveillance. Moreover, surgeons identified CM-SSI not captured under current NSQIP definitions. Future efforts to expand CM-SSI identification may lead to improved and potentially automated approaches to survey for postoperative SSI in clinical practice.

KW - Bayesian network

KW - Colorectal surgery

KW - Natural language processing

KW - Surgical site infection

UR - http://www.scopus.com/inward/record.url?scp=84994832127&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84994832127&partnerID=8YFLogxK

U2 - 10.1016/j.jss.2016.09.058

DO - 10.1016/j.jss.2016.09.058

M3 - Article

VL - 209

SP - 168

EP - 173

JO - Journal of Surgical Research

JF - Journal of Surgical Research

SN - 0022-4804

ER -