TY - JOUR
T1 - A pilot study to predict cardiac arrest in the pediatric intensive care unit
AU - Kenet, Adam L.
AU - Pemmaraju, Rahul
AU - Ghate, Sejal
AU - Raghunath, Shreeya
AU - Zhang, Yifan
AU - Yuan, Mordred
AU - Wei, Tony Y.
AU - Desman, Jacob M.
AU - Greenstein, Joseph L.
AU - Taylor, Casey O.
AU - Ruchti, Timothy
AU - Fackler, James
AU - Bergmann, Jules
N1 - Funding Information:
We gratefully acknowledge the work of Dr. Elizabeth Hunt and Dr. Jordan Duval-Arnould who curated the data used in these models and assured that IHCAs were labeled at onset. These labels were crucial to the success of our work.
Publisher Copyright:
© 2023 Elsevier B.V.
PY - 2023/4
Y1 - 2023/4
N2 - 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.
AB - 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.
KW - Biomedical engineering
KW - Cardiac arrest
KW - Computational medicine
KW - Critical care medicine
KW - Heart rate variability
KW - High frequency waveform data
KW - Machine learning
KW - Pediatric intensive care unit
KW - Predictive modeling
UR - http://www.scopus.com/inward/record.url?scp=85149199930&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85149199930&partnerID=8YFLogxK
U2 - 10.1016/j.resuscitation.2023.109740
DO - 10.1016/j.resuscitation.2023.109740
M3 - Article
C2 - 36805101
AN - SCOPUS:85149199930
SN - 0300-9572
VL - 185
JO - Resuscitation
JF - Resuscitation
M1 - 109740
ER -