Automated machine learning-based model predicts postoperative delirium using readily extractable perioperative collected electronic data

Xiao-Yi Hu; He Liu; Xue Zhao; Xun Sun; Jian Zhou; Xing Gao; Hui-Lian Guan; Yang Zhou; Qiu Zhao; Yuan Han; Jun-Li Cao

doi:10.1111/cns.13758

Automated machine learning-based model predicts postoperative delirium using readily extractable perioperative collected electronic data

CNS Neurosci Ther. 2022 Apr;28(4):608-618. doi: 10.1111/cns.13758. Epub 2021 Nov 18.

Authors

Xiao-Yi Hu^#^{1

2}, He Liu^#³, Xue Zhao¹, Xun Sun^{1

2}, Jian Zhou^{1

2}, Xing Gao^{2

4}, Hui-Lian Guan^{2

5}, Yang Zhou², Qiu Zhao^{1

2}, Yuan Han⁶, Jun-Li Cao^{1

2}

Affiliations

¹ Department of Anesthesiology, The Affiliated Hospital of Xuzhou Medical University, Jiangsu Province, Xuzhou City, China.
² Jiangsu Province Key Laboratory of Anesthesiology & NMPA Key Laboratory for Research and Evaluation of Narcotic and Psychotropic Drugs, Xuzhou Medical University, Jiangsu Province, Xuzhou City, China.
³ Department of Anesthesiology, The Affiliated Huzhou Hospital, Zhejiang University School of Medicine, Huzhou Central Hospital, Zhejiang Province, Huzhou City, China.
⁴ Department of Anesthesiology, Changzhou First People's Hospital, Changzhou, Jiangsu, China.
⁵ Department of Anesthesiology, The First Affiliated Hospital of Bengbu Medical College, Bengbu, Anhui, China.
⁶ Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China.

^# Contributed equally.

Abstract

Objective: Postoperative delirium (POD) is a common postoperative complication that is relevant to poor outcomes. Therefore, it is critical to find effective methods to identify patients with high risk of POD rapidly. Creating a fully automated score based on an automated machine-learning algorithm may be a method to predict the incidence of POD quickly.

Materials and methods: This is the secondary analysis of an observational study, including 531 surgical patients who underwent general anesthesia. The least absolute shrinkage and selection operator (LASSO) was used to screen essential features associated with POD. Finally, eight features (age, intraoperative blood loss, anesthesia duration, extubation time, intensive care unit [ICU] admission, mini-mental state examination score [MMSE], Charlson comorbidity index [CCI], postoperative neutrophil-to-lymphocyte ratio [NLR]) were used to established models. Four models, logistic regression, random forest, extreme gradient boosted trees, and support vector machines, were built in a training set (70% of participants) and evaluated in the remaining testing sample (30% of participants). Multivariate logistic regression analysis was used to explore independent risk factors for POD further.

Results: Model 1 (logistic regression model) was found to outperform other classifier models in testing data (area under the curve [AUC] of 80.44%, 95% confidence interval [CI] 72.24%-88.64%) and achieve the lowest Brier Score as well. These variables including age (OR = 1.054, 95%CI: 1.017~1.093), extubation time (OR = 1.027, 95%CI: 1.012~1.044), ICU admission (OR = 2.238, 95%CI: 1.313~3.793), MMSE (OR = 0.929, 95%CI: 0.876~0.984), CCI (OR = 1.197, 95%CI: 1.038~1.384), and postoperative NLR (OR = 1.029, 95%CI: 1.002~1.057) were independent risk factors for POD in this study.

Conclusions: We have built and validated a high-performing algorithm to demonstrate the extent to which patient risk changes of POD during the perioperative period, thus leading to a rational therapeutic choice.

Keywords: delirium; machine learning; model prediction; nomogram; postoperative.

Publication types

Observational Study
Research Support, Non-U.S. Gov't

MeSH terms

Delirium* / diagnosis
Delirium* / epidemiology
Delirium* / etiology
Electronics
Humans
Machine Learning
Postoperative Complications / diagnosis
Postoperative Complications / epidemiology
Retrospective Studies
Risk Factors