Accepted for/Published in: JMIR Medical Informatics
Date Submitted: Aug 20, 2019
Open Peer Review Period: Aug 20, 2019 - Aug 28, 2019
Date Accepted: Jul 14, 2020
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Machine learning prediction of cardiac arrest in emergency department: sequential characteristics for clinical validity
ABSTRACT
Background:
The development and application of clinical prediction models using machine learning in clinical decision support systems has attracted increasing attention.
Objective:
The aim of this study is to develop a prediction model for cardiac arrest using machine learning and sequential characteristics in emergency department (ED) and to perform validations for clinical usefulness.
Methods:
This retrospective study was conducted for ED patients from a tertiary academic hospital that suffered cardiac arrest. To resolve the class imbalance problem, sampling was performed using propensity score matching. The dataset was randomly allocated to the development cohort (80%) and the validation cohort (20%). We trained three machine learning algorithms with repeated 10-fold cross-validation.
Results:
The main performance parameters were the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPRC). The random forest algorithm (AUROC = 0.97; AUPRC = 0.85) outperformed the recurrent neural network (AUROC = 0.96; AUPRC = 0.80) and the logistic regression algorithm (AUROC=0.92; AUPRC=0.72). The model performance over time was maintained with AUROC of at least 80% across monitoring time points during 24 hours before event occurrence.
Conclusions:
We developed a cardiac arrest prediction model using machine learning and sequential characteristics in ED. The model was validated for clinical usefulness using chronological visualisation focused on clinical usability.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.