Abstract

Background. Perioperative hypotension is associated with adverse outcomes in patients undergoing surgery. A computer-based model that integrates related factors and predicts the risk of hypotension would be helpful in clinical anesthesia. The purpose of this study was to develop artificial neural network (ANN) models to identify patients at high risk for postinduction hypotension during general anesthesia. Methods. Anesthesia records for March through November 2007 were reviewed, and 1017 records were analyzed. Eleven patient-related, 2 surgical, and 5 anesthetic variables were used to develop the ANN and logistic regression (LR) models. The quality of the models was evaluated by an external validation data set. Three clinicians were asked to make predictions of the same validation data set on a case-by-case basis. Results. The ANN model had an accuracy of 82.3%, sensitivity of 76.4%, and specificity of 85.6%. The accuracy of the LR model was 76.5%, the sensitivity was 74.5%, and specificity was 77.7%. The area under the receiver operating characteristic curve for the ANN and LR models was 0.893 and 0.840. The clinicians had the lowest predictive accuracy and sensitivity compared with the ANN and LR models. Conclusions. The ANN model developed in this study had good discrimination and calibration and would provide decision support to clinicians and increase vigilance for patients at high risk of postinduction hypotension during general anesthesia.

Original languageEnglish
Pages (from-to)308-314
Number of pages7
JournalMedical Decision Making
Volume31
Issue number2
DOIs
Publication statusPublished - Mar 2011

Fingerprint

Hypotension
General Anesthesia
Logistic Models
Neural Networks (Computer)
Anesthesia
ROC Curve
Computer Simulation
Calibration
Anesthetics
Sensitivity and Specificity

Keywords

  • Anesthesiology
  • Artificial neural networks
  • Logistic regression models
  • ROC curve analysis

ASJC Scopus subject areas

  • Health Policy
  • Medicine(all)

Cite this

@article{0704a13dcb444e8c906fe87b2410bbaa,
title = "Application of an artificial neural network to predict postinduction hypotension during general anesthesia",
abstract = "Background. Perioperative hypotension is associated with adverse outcomes in patients undergoing surgery. A computer-based model that integrates related factors and predicts the risk of hypotension would be helpful in clinical anesthesia. The purpose of this study was to develop artificial neural network (ANN) models to identify patients at high risk for postinduction hypotension during general anesthesia. Methods. Anesthesia records for March through November 2007 were reviewed, and 1017 records were analyzed. Eleven patient-related, 2 surgical, and 5 anesthetic variables were used to develop the ANN and logistic regression (LR) models. The quality of the models was evaluated by an external validation data set. Three clinicians were asked to make predictions of the same validation data set on a case-by-case basis. Results. The ANN model had an accuracy of 82.3{\%}, sensitivity of 76.4{\%}, and specificity of 85.6{\%}. The accuracy of the LR model was 76.5{\%}, the sensitivity was 74.5{\%}, and specificity was 77.7{\%}. The area under the receiver operating characteristic curve for the ANN and LR models was 0.893 and 0.840. The clinicians had the lowest predictive accuracy and sensitivity compared with the ANN and LR models. Conclusions. The ANN model developed in this study had good discrimination and calibration and would provide decision support to clinicians and increase vigilance for patients at high risk of postinduction hypotension during general anesthesia.",
keywords = "Anesthesiology, Artificial neural networks, Logistic regression models, ROC curve analysis",
author = "Chao-Shun Lin and Chuen-Chau Chang and Chiu, {Jainn Shiun} and Yuan-Wen Lee and Jui-An Lin and Mok, {Martin S.} and Hung-Wen Chiu and Yu-Chuan Li",
year = "2011",
month = "3",
doi = "10.1177/0272989X10379648",
language = "English",
volume = "31",
pages = "308--314",
journal = "Medical Decision Making",
issn = "0272-989X",
publisher = "SAGE Publications Inc.",
number = "2",

}

TY - JOUR

T1 - Application of an artificial neural network to predict postinduction hypotension during general anesthesia

AU - Lin, Chao-Shun

AU - Chang, Chuen-Chau

AU - Chiu, Jainn Shiun

AU - Lee, Yuan-Wen

AU - Lin, Jui-An

AU - Mok, Martin S.

AU - Chiu, Hung-Wen

AU - Li, Yu-Chuan

PY - 2011/3

Y1 - 2011/3

N2 - Background. Perioperative hypotension is associated with adverse outcomes in patients undergoing surgery. A computer-based model that integrates related factors and predicts the risk of hypotension would be helpful in clinical anesthesia. The purpose of this study was to develop artificial neural network (ANN) models to identify patients at high risk for postinduction hypotension during general anesthesia. Methods. Anesthesia records for March through November 2007 were reviewed, and 1017 records were analyzed. Eleven patient-related, 2 surgical, and 5 anesthetic variables were used to develop the ANN and logistic regression (LR) models. The quality of the models was evaluated by an external validation data set. Three clinicians were asked to make predictions of the same validation data set on a case-by-case basis. Results. The ANN model had an accuracy of 82.3%, sensitivity of 76.4%, and specificity of 85.6%. The accuracy of the LR model was 76.5%, the sensitivity was 74.5%, and specificity was 77.7%. The area under the receiver operating characteristic curve for the ANN and LR models was 0.893 and 0.840. The clinicians had the lowest predictive accuracy and sensitivity compared with the ANN and LR models. Conclusions. The ANN model developed in this study had good discrimination and calibration and would provide decision support to clinicians and increase vigilance for patients at high risk of postinduction hypotension during general anesthesia.

AB - Background. Perioperative hypotension is associated with adverse outcomes in patients undergoing surgery. A computer-based model that integrates related factors and predicts the risk of hypotension would be helpful in clinical anesthesia. The purpose of this study was to develop artificial neural network (ANN) models to identify patients at high risk for postinduction hypotension during general anesthesia. Methods. Anesthesia records for March through November 2007 were reviewed, and 1017 records were analyzed. Eleven patient-related, 2 surgical, and 5 anesthetic variables were used to develop the ANN and logistic regression (LR) models. The quality of the models was evaluated by an external validation data set. Three clinicians were asked to make predictions of the same validation data set on a case-by-case basis. Results. The ANN model had an accuracy of 82.3%, sensitivity of 76.4%, and specificity of 85.6%. The accuracy of the LR model was 76.5%, the sensitivity was 74.5%, and specificity was 77.7%. The area under the receiver operating characteristic curve for the ANN and LR models was 0.893 and 0.840. The clinicians had the lowest predictive accuracy and sensitivity compared with the ANN and LR models. Conclusions. The ANN model developed in this study had good discrimination and calibration and would provide decision support to clinicians and increase vigilance for patients at high risk of postinduction hypotension during general anesthesia.

KW - Anesthesiology

KW - Artificial neural networks

KW - Logistic regression models

KW - ROC curve analysis

UR - http://www.scopus.com/inward/record.url?scp=79953836782&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79953836782&partnerID=8YFLogxK

U2 - 10.1177/0272989X10379648

DO - 10.1177/0272989X10379648

M3 - Article

C2 - 20876347

AN - SCOPUS:79953836782

VL - 31

SP - 308

EP - 314

JO - Medical Decision Making

JF - Medical Decision Making

SN - 0272-989X

IS - 2

ER -