A pilot study to predict cardiac arrest in the pediatric intensive care unit.

Biomedical engineering Cardiac arrest Computational medicine Critical care medicine Heart rate variability High frequency waveform data Machine learning Pediatric intensive care unit Predictive modeling

Journal

Resuscitation
ISSN: 1873-1570
Titre abrégé: Resuscitation
Pays: Ireland
ID NLM: 0332173

Informations de publication

Date de publication:
04 2023
Historique:
received: 08 11 2022
revised: 09 02 2023
accepted: 10 02 2023
medline: 31 3 2023
pubmed: 23 2 2023
entrez: 22 2 2023
Statut: ppublish

Résumé

Cardiac arrest is a leading cause of mortality prior to discharge for children admitted to the pediatric intensive care unit. To address this problem, we used machine learning to predict cardiac arrest up to three hours in advance. Our data consists of 240 Hz ECG waveform data, 0.5 Hz physiological time series data, medications, and demographics from 1,145 patients in the pediatric intensive care unit at the Johns Hopkins Hospital, 15 of whom experienced a cardiac arrest. The data were divided into training, validating, and testing sets, and features were generated every five minutes. 23 heart rate variability (HRV) metrics were determined from ECG waveforms. 96 summary statistics were calculated for 12 vital signs, such as respiratory rate and blood pressure. Medications were classified into 42 therapeutic drug classes. Binary features were generated to indicate the administration of these different drugs. Next, six machine learning models were evaluated: logistic regression, support vector machine, random forest, XGBoost, LightGBM, and a soft voting ensemble. XGBoost performed the best, with 0.971 auROC, 0.797 auPRC, 99.5% sensitivity, and 69.6% specificity on an independent test set. We have created high-performing models that identify signatures of in-hospital cardiac arrest (IHCA) that may not be evident to clinicians. These signatures include a combination of heart rate variability metrics, vital signs data, and therapeutic drug classes. These machine learning models can predict IHCA up to three hours prior to onset with high performance, allowing clinicians to intervene earlier, improving patient outcomes.

Sections du résumé

BACKGROUND
Cardiac arrest is a leading cause of mortality prior to discharge for children admitted to the pediatric intensive care unit. To address this problem, we used machine learning to predict cardiac arrest up to three hours in advance.
METHODS
Our data consists of 240 Hz ECG waveform data, 0.5 Hz physiological time series data, medications, and demographics from 1,145 patients in the pediatric intensive care unit at the Johns Hopkins Hospital, 15 of whom experienced a cardiac arrest. The data were divided into training, validating, and testing sets, and features were generated every five minutes. 23 heart rate variability (HRV) metrics were determined from ECG waveforms. 96 summary statistics were calculated for 12 vital signs, such as respiratory rate and blood pressure. Medications were classified into 42 therapeutic drug classes. Binary features were generated to indicate the administration of these different drugs. Next, six machine learning models were evaluated: logistic regression, support vector machine, random forest, XGBoost, LightGBM, and a soft voting ensemble.
RESULTS
XGBoost performed the best, with 0.971 auROC, 0.797 auPRC, 99.5% sensitivity, and 69.6% specificity on an independent test set.
CONCLUSION
We have created high-performing models that identify signatures of in-hospital cardiac arrest (IHCA) that may not be evident to clinicians. These signatures include a combination of heart rate variability metrics, vital signs data, and therapeutic drug classes. These machine learning models can predict IHCA up to three hours prior to onset with high performance, allowing clinicians to intervene earlier, improving patient outcomes.

Identifiants

pubmed: 36805101
pii: S0300-9572(23)00053-9
doi: 10.1016/j.resuscitation.2023.109740
pii:
doi:

Types de publication

Journal Article Research Support, Non-U.S. Gov't

Langues

eng

Sous-ensembles de citation

IM

Pagination

109740

Informations de copyright

Copyright © 2023 Elsevier B.V. All rights reserved.

Auteurs

Adam L Kenet (AL)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States. Electronic address: akenet1@jhu.edu.

Rahul Pemmaraju (R)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Sejal Ghate (S)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Shreeya Raghunath (S)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Yifan Zhang (Y)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Mordred Yuan (M)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Tony Y Wei (TY)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Jacob M Desman (JM)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Joseph L Greenstein (JL)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States.

Casey O Taylor (CO)

Department of Biomedical Engineering, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Institute for Computational Medicine, Johns Hopkins University Whiting School of Engineering, Baltimore, MD, United States; Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, United States.

Timothy Ruchti (T)

Nihon Kohden Digital Health Solutions Inc, Irvine, CA, United States.

James Fackler (J)

Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, United States.

Jules Bergmann (J)

Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, United States.

Articles similaires

[Redispensing of expensive oral anticancer medicines: a practical application].

Lisanne N van Merendonk, Kübra Akgöl, Bastiaan Nuijen
1.00
Humans Antineoplastic Agents Administration, Oral Drug Costs Counterfeit Drugs

Smoking Cessation and Incident Cardiovascular Disease.

Jun Hwan Cho, Seung Yong Shin, Hoseob Kim et al.
1.00
Humans Male Smoking Cessation Cardiovascular Diseases Female
Humans United States Aged Cross-Sectional Studies Medicare Part C
1.00
Humans Yoga Low Back Pain Female Male

Classifications MeSH