Prediction of sepsis patients using machine learning approach: A meta-analysis

Md Mohaimenul Islam, Tahmina Nasrin, Bruno Andreas Walther, Chieh Chen Wu, Hsuan Chia Yang, Yu Chuan Li

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Study objective: Sepsis is a common and major health crisis in hospitals globally. An innovative and feasible tool for predicting sepsis remains elusive. However, early and accurate prediction of sepsis could help physicians with proper treatments and minimize the diagnostic uncertainty. Machine learning models could help to identify potential clinical variables and provide higher performance than existing traditional low-performance models. We therefore performed a meta-analysis of observational studies to quantify the performance of a machine learning model to predict sepsis. Methods: A comprehensive literature search was conducted through the electronic database (e.g. PubMed, Scopus, Google Scholar, EMBASE, etc.) between January 1, 2000, and March 1, 2018. All the studies published in English and reporting the sepsis prediction using machine learning algorithms were considered in this study. Two authors independently extracted valuable information from the included studies. Inclusion and exclusion of studies were based on the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines. Results: A total of 7 out of 135 studies met all of our inclusion criteria. For machine learning models, the pooled area under receiving operating curve (SAUROC) for predicting sepsis onset 3 to 4 h before, was 0.89 (95%CI: 0.86–0.92); sensitivity 0.81 (95%CI:0.80–0.81), and specificity 0.72 (95%CI:0.72–0.72) whereas the pooled SAUROC for SIRS, MEWS, and SOFA was 0.70, 0.50, and 0.78. Additionally, diagnostic odd ratio for machine learning, SIRS, MEWS, and SOFA was 15.17 (95%CI: 9.51–24.20), 3.23 (95%CI: 1.52–6.87), 31.99 (95% CI: 1.54–666.74), and 3.75(95%CI: 2.06–6.83). Conclusion: Our study findings suggest that the machine learning approach had a better performance than the existing sepsis scoring systems in predicting sepsis.

Original languageEnglish
Pages (from-to)1-9
Number of pages9
JournalComputer Methods and Programs in Biomedicine
Volume170
DOIs
Publication statusPublished - Mar 1 2019

Fingerprint

Learning systems
Meta-Analysis
Sepsis
Learning algorithms
Machine Learning
PubMed
Health
Uncertainty
Observational Studies
Odds Ratio
Databases
Guidelines
Physicians

Keywords

  • Area under receiver operating curve
  • Diagnostic odd ratio
  • Machine learning
  • Sepsis

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Health Informatics

Cite this

Prediction of sepsis patients using machine learning approach : A meta-analysis. / Islam, Md Mohaimenul; Nasrin, Tahmina; Walther, Bruno Andreas; Wu, Chieh Chen; Yang, Hsuan Chia; Li, Yu Chuan.

In: Computer Methods and Programs in Biomedicine, Vol. 170, 01.03.2019, p. 1-9.

Research output: Contribution to journalArticle

Islam, Md Mohaimenul ; Nasrin, Tahmina ; Walther, Bruno Andreas ; Wu, Chieh Chen ; Yang, Hsuan Chia ; Li, Yu Chuan. / Prediction of sepsis patients using machine learning approach : A meta-analysis. In: Computer Methods and Programs in Biomedicine. 2019 ; Vol. 170. pp. 1-9.
@article{03d317503ad6457d9f5738ec992d6645,
title = "Prediction of sepsis patients using machine learning approach: A meta-analysis",
abstract = "Study objective: Sepsis is a common and major health crisis in hospitals globally. An innovative and feasible tool for predicting sepsis remains elusive. However, early and accurate prediction of sepsis could help physicians with proper treatments and minimize the diagnostic uncertainty. Machine learning models could help to identify potential clinical variables and provide higher performance than existing traditional low-performance models. We therefore performed a meta-analysis of observational studies to quantify the performance of a machine learning model to predict sepsis. Methods: A comprehensive literature search was conducted through the electronic database (e.g. PubMed, Scopus, Google Scholar, EMBASE, etc.) between January 1, 2000, and March 1, 2018. All the studies published in English and reporting the sepsis prediction using machine learning algorithms were considered in this study. Two authors independently extracted valuable information from the included studies. Inclusion and exclusion of studies were based on the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines. Results: A total of 7 out of 135 studies met all of our inclusion criteria. For machine learning models, the pooled area under receiving operating curve (SAUROC) for predicting sepsis onset 3 to 4 h before, was 0.89 (95{\%}CI: 0.86–0.92); sensitivity 0.81 (95{\%}CI:0.80–0.81), and specificity 0.72 (95{\%}CI:0.72–0.72) whereas the pooled SAUROC for SIRS, MEWS, and SOFA was 0.70, 0.50, and 0.78. Additionally, diagnostic odd ratio for machine learning, SIRS, MEWS, and SOFA was 15.17 (95{\%}CI: 9.51–24.20), 3.23 (95{\%}CI: 1.52–6.87), 31.99 (95{\%} CI: 1.54–666.74), and 3.75(95{\%}CI: 2.06–6.83). Conclusion: Our study findings suggest that the machine learning approach had a better performance than the existing sepsis scoring systems in predicting sepsis.",
keywords = "Area under receiver operating curve, Diagnostic odd ratio, Machine learning, Sepsis",
author = "Islam, {Md Mohaimenul} and Tahmina Nasrin and Walther, {Bruno Andreas} and Wu, {Chieh Chen} and Yang, {Hsuan Chia} and Li, {Yu Chuan}",
year = "2019",
month = "3",
day = "1",
doi = "10.1016/j.cmpb.2018.12.027",
language = "English",
volume = "170",
pages = "1--9",
journal = "Computer Methods and Programs in Biomedicine",
issn = "0169-2607",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - Prediction of sepsis patients using machine learning approach

T2 - A meta-analysis

AU - Islam, Md Mohaimenul

AU - Nasrin, Tahmina

AU - Walther, Bruno Andreas

AU - Wu, Chieh Chen

AU - Yang, Hsuan Chia

AU - Li, Yu Chuan

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Study objective: Sepsis is a common and major health crisis in hospitals globally. An innovative and feasible tool for predicting sepsis remains elusive. However, early and accurate prediction of sepsis could help physicians with proper treatments and minimize the diagnostic uncertainty. Machine learning models could help to identify potential clinical variables and provide higher performance than existing traditional low-performance models. We therefore performed a meta-analysis of observational studies to quantify the performance of a machine learning model to predict sepsis. Methods: A comprehensive literature search was conducted through the electronic database (e.g. PubMed, Scopus, Google Scholar, EMBASE, etc.) between January 1, 2000, and March 1, 2018. All the studies published in English and reporting the sepsis prediction using machine learning algorithms were considered in this study. Two authors independently extracted valuable information from the included studies. Inclusion and exclusion of studies were based on the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines. Results: A total of 7 out of 135 studies met all of our inclusion criteria. For machine learning models, the pooled area under receiving operating curve (SAUROC) for predicting sepsis onset 3 to 4 h before, was 0.89 (95%CI: 0.86–0.92); sensitivity 0.81 (95%CI:0.80–0.81), and specificity 0.72 (95%CI:0.72–0.72) whereas the pooled SAUROC for SIRS, MEWS, and SOFA was 0.70, 0.50, and 0.78. Additionally, diagnostic odd ratio for machine learning, SIRS, MEWS, and SOFA was 15.17 (95%CI: 9.51–24.20), 3.23 (95%CI: 1.52–6.87), 31.99 (95% CI: 1.54–666.74), and 3.75(95%CI: 2.06–6.83). Conclusion: Our study findings suggest that the machine learning approach had a better performance than the existing sepsis scoring systems in predicting sepsis.

AB - Study objective: Sepsis is a common and major health crisis in hospitals globally. An innovative and feasible tool for predicting sepsis remains elusive. However, early and accurate prediction of sepsis could help physicians with proper treatments and minimize the diagnostic uncertainty. Machine learning models could help to identify potential clinical variables and provide higher performance than existing traditional low-performance models. We therefore performed a meta-analysis of observational studies to quantify the performance of a machine learning model to predict sepsis. Methods: A comprehensive literature search was conducted through the electronic database (e.g. PubMed, Scopus, Google Scholar, EMBASE, etc.) between January 1, 2000, and March 1, 2018. All the studies published in English and reporting the sepsis prediction using machine learning algorithms were considered in this study. Two authors independently extracted valuable information from the included studies. Inclusion and exclusion of studies were based on the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines. Results: A total of 7 out of 135 studies met all of our inclusion criteria. For machine learning models, the pooled area under receiving operating curve (SAUROC) for predicting sepsis onset 3 to 4 h before, was 0.89 (95%CI: 0.86–0.92); sensitivity 0.81 (95%CI:0.80–0.81), and specificity 0.72 (95%CI:0.72–0.72) whereas the pooled SAUROC for SIRS, MEWS, and SOFA was 0.70, 0.50, and 0.78. Additionally, diagnostic odd ratio for machine learning, SIRS, MEWS, and SOFA was 15.17 (95%CI: 9.51–24.20), 3.23 (95%CI: 1.52–6.87), 31.99 (95% CI: 1.54–666.74), and 3.75(95%CI: 2.06–6.83). Conclusion: Our study findings suggest that the machine learning approach had a better performance than the existing sepsis scoring systems in predicting sepsis.

KW - Area under receiver operating curve

KW - Diagnostic odd ratio

KW - Machine learning

KW - Sepsis

UR - http://www.scopus.com/inward/record.url?scp=85059549180&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059549180&partnerID=8YFLogxK

U2 - 10.1016/j.cmpb.2018.12.027

DO - 10.1016/j.cmpb.2018.12.027

M3 - Article

C2 - 30712598

AN - SCOPUS:85059549180

VL - 170

SP - 1

EP - 9

JO - Computer Methods and Programs in Biomedicine

JF - Computer Methods and Programs in Biomedicine

SN - 0169-2607

ER -