Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement

Joshua Parreco, Antonio Hidalgo, Jonathan J. Parks, Robert Kozol, Rishi Rattan

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Background: Early identification of critically ill patients who will require prolonged mechanical ventilation (PMV) has proven to be difficult. The purpose of this study was to use machine learning to identify patients at risk for PMV and tracheostomy placement. Materials and methods: The Multiparameter Intelligent Monitoring in Intensive Care III database was queried for all intensive care unit (ICU) stays with mechanical ventilation. PMV was defined as ventilation >7 d. Classifiers with a gradient-boosted decision trees algorithm were created for the outcomes of PMV and tracheostomy placement. The variables used were six different severity-of-illness scores calculated on the first day of ICU admission including their components and 30 comorbidities. Mean receiver operating characteristic curves were calculated for the outcomes, and variable importance was quantified. Results: There were 20,262 ICU stays identified. PMV was required in 13.6%, and tracheostomy was performed in 6.6% of patients. The classifier for predicting PMV was able to achieve a mean area under the curve (AUC) of 0.820 ± 0.016, and tracheostomy was predicted with an AUC of 0.830 ± 0.011. There were 60.7% patients admitted to a surgical ICU, and the classifiers for these patients predicted PMV with an AUC of 0.852 ± 0.017 and tracheostomy with an AUC of 0.869 ± 0.015. The variable with the highest importance for predicting PMV was the logistic organ dysfunction score pulmonary component (13%), and the most important comorbidity in predicting tracheostomy was cardiac arrhythmia (12%). Conclusions: This study demonstrates the use of artificial intelligence through machine-learning classifiers for the early identification of patients at risk for PMV and tracheostomy. Application of these identification techniques could lead to improved outcomes by allowing for early intervention.

Original languageEnglish (US)
Pages (from-to)179-187
Number of pages9
JournalJournal of Surgical Research
Volume228
DOIs
StatePublished - Aug 1 2018

Fingerprint

Tracheostomy
Artificial Intelligence
Artificial Respiration
Area Under Curve
Intensive Care Units
Critical Care
Comorbidity
Organ Dysfunction Scores
Decision Trees
Critical Illness
ROC Curve
Ventilation
Cardiac Arrhythmias
Databases
Lung

Keywords

  • Artificial intelligence
  • Critical care
  • Machine learning
  • Prolonged mechanical ventilation
  • Tracheostomy

ASJC Scopus subject areas

  • Surgery

Cite this

Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement. / Parreco, Joshua; Hidalgo, Antonio; Parks, Jonathan J.; Kozol, Robert; Rattan, Rishi.

In: Journal of Surgical Research, Vol. 228, 01.08.2018, p. 179-187.

Research output: Contribution to journalArticle

Parreco, Joshua ; Hidalgo, Antonio ; Parks, Jonathan J. ; Kozol, Robert ; Rattan, Rishi. / Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement. In: Journal of Surgical Research. 2018 ; Vol. 228. pp. 179-187.
@article{500d0c1f5de04d3099b2d363d92da01c,
title = "Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement",
abstract = "Background: Early identification of critically ill patients who will require prolonged mechanical ventilation (PMV) has proven to be difficult. The purpose of this study was to use machine learning to identify patients at risk for PMV and tracheostomy placement. Materials and methods: The Multiparameter Intelligent Monitoring in Intensive Care III database was queried for all intensive care unit (ICU) stays with mechanical ventilation. PMV was defined as ventilation >7 d. Classifiers with a gradient-boosted decision trees algorithm were created for the outcomes of PMV and tracheostomy placement. The variables used were six different severity-of-illness scores calculated on the first day of ICU admission including their components and 30 comorbidities. Mean receiver operating characteristic curves were calculated for the outcomes, and variable importance was quantified. Results: There were 20,262 ICU stays identified. PMV was required in 13.6{\%}, and tracheostomy was performed in 6.6{\%} of patients. The classifier for predicting PMV was able to achieve a mean area under the curve (AUC) of 0.820 ± 0.016, and tracheostomy was predicted with an AUC of 0.830 ± 0.011. There were 60.7{\%} patients admitted to a surgical ICU, and the classifiers for these patients predicted PMV with an AUC of 0.852 ± 0.017 and tracheostomy with an AUC of 0.869 ± 0.015. The variable with the highest importance for predicting PMV was the logistic organ dysfunction score pulmonary component (13{\%}), and the most important comorbidity in predicting tracheostomy was cardiac arrhythmia (12{\%}). Conclusions: This study demonstrates the use of artificial intelligence through machine-learning classifiers for the early identification of patients at risk for PMV and tracheostomy. Application of these identification techniques could lead to improved outcomes by allowing for early intervention.",
keywords = "Artificial intelligence, Critical care, Machine learning, Prolonged mechanical ventilation, Tracheostomy",
author = "Joshua Parreco and Antonio Hidalgo and Parks, {Jonathan J.} and Robert Kozol and Rishi Rattan",
year = "2018",
month = "8",
day = "1",
doi = "10.1016/j.jss.2018.03.028",
language = "English (US)",
volume = "228",
pages = "179--187",
journal = "Journal of Surgical Research",
issn = "0022-4804",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement

AU - Parreco, Joshua

AU - Hidalgo, Antonio

AU - Parks, Jonathan J.

AU - Kozol, Robert

AU - Rattan, Rishi

PY - 2018/8/1

Y1 - 2018/8/1

N2 - Background: Early identification of critically ill patients who will require prolonged mechanical ventilation (PMV) has proven to be difficult. The purpose of this study was to use machine learning to identify patients at risk for PMV and tracheostomy placement. Materials and methods: The Multiparameter Intelligent Monitoring in Intensive Care III database was queried for all intensive care unit (ICU) stays with mechanical ventilation. PMV was defined as ventilation >7 d. Classifiers with a gradient-boosted decision trees algorithm were created for the outcomes of PMV and tracheostomy placement. The variables used were six different severity-of-illness scores calculated on the first day of ICU admission including their components and 30 comorbidities. Mean receiver operating characteristic curves were calculated for the outcomes, and variable importance was quantified. Results: There were 20,262 ICU stays identified. PMV was required in 13.6%, and tracheostomy was performed in 6.6% of patients. The classifier for predicting PMV was able to achieve a mean area under the curve (AUC) of 0.820 ± 0.016, and tracheostomy was predicted with an AUC of 0.830 ± 0.011. There were 60.7% patients admitted to a surgical ICU, and the classifiers for these patients predicted PMV with an AUC of 0.852 ± 0.017 and tracheostomy with an AUC of 0.869 ± 0.015. The variable with the highest importance for predicting PMV was the logistic organ dysfunction score pulmonary component (13%), and the most important comorbidity in predicting tracheostomy was cardiac arrhythmia (12%). Conclusions: This study demonstrates the use of artificial intelligence through machine-learning classifiers for the early identification of patients at risk for PMV and tracheostomy. Application of these identification techniques could lead to improved outcomes by allowing for early intervention.

AB - Background: Early identification of critically ill patients who will require prolonged mechanical ventilation (PMV) has proven to be difficult. The purpose of this study was to use machine learning to identify patients at risk for PMV and tracheostomy placement. Materials and methods: The Multiparameter Intelligent Monitoring in Intensive Care III database was queried for all intensive care unit (ICU) stays with mechanical ventilation. PMV was defined as ventilation >7 d. Classifiers with a gradient-boosted decision trees algorithm were created for the outcomes of PMV and tracheostomy placement. The variables used were six different severity-of-illness scores calculated on the first day of ICU admission including their components and 30 comorbidities. Mean receiver operating characteristic curves were calculated for the outcomes, and variable importance was quantified. Results: There were 20,262 ICU stays identified. PMV was required in 13.6%, and tracheostomy was performed in 6.6% of patients. The classifier for predicting PMV was able to achieve a mean area under the curve (AUC) of 0.820 ± 0.016, and tracheostomy was predicted with an AUC of 0.830 ± 0.011. There were 60.7% patients admitted to a surgical ICU, and the classifiers for these patients predicted PMV with an AUC of 0.852 ± 0.017 and tracheostomy with an AUC of 0.869 ± 0.015. The variable with the highest importance for predicting PMV was the logistic organ dysfunction score pulmonary component (13%), and the most important comorbidity in predicting tracheostomy was cardiac arrhythmia (12%). Conclusions: This study demonstrates the use of artificial intelligence through machine-learning classifiers for the early identification of patients at risk for PMV and tracheostomy. Application of these identification techniques could lead to improved outcomes by allowing for early intervention.

KW - Artificial intelligence

KW - Critical care

KW - Machine learning

KW - Prolonged mechanical ventilation

KW - Tracheostomy

UR - http://www.scopus.com/inward/record.url?scp=85045065764&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045065764&partnerID=8YFLogxK

U2 - 10.1016/j.jss.2018.03.028

DO - 10.1016/j.jss.2018.03.028

M3 - Article

AN - SCOPUS:85045065764

VL - 228

SP - 179

EP - 187

JO - Journal of Surgical Research

JF - Journal of Surgical Research

SN - 0022-4804

ER -