Development of a machine-learning algorithm to predict in-hospital cardiac arrest for emergency department patients using a nationwide database

Abstract

In this retrospective observational study, we aimed to develop a machine-learning model using data obtained at the prehospital stage to predict in-hospital cardiac arrest in the emergency department (ED) of patients transferred via emergency medical services. The dataset was constructed by attaching the prehospital information from the National Fire Agency and hospital factors to data from the National Emergency Department Information System. Machine-learning models were developed using patient variables, with and without hospital factors. We validated model performance and used the SHapley Additive exPlanation model interpretation. In-hospital cardiac arrest occurred in 5431 of the 1,350,693 patients (0.4%). The extreme gradient boosting model showed the best performance with area under receiver operating curve of 0.9267 when incorporating the hospital factor. Oxygen supply, age, oxygen saturation, systolic blood pressure, the number of ED beds, ED occupancy, and pulse rate were the most influential variables, in that order. ED occupancy and in-hospital cardiac arrest occurrence were positively correlated, and the impact of ED occupancy appeared greater in small hospitals. The machine-learning predictive model using the integrated information acquired in the prehospital stage effectively predicted in-hospital cardiac arrest in the ED and can contribute to the efficient operation of emergency medical systems.

Development of a Machine-Learning Algorithm for Predicting the Arrival Time and Assessing Factors Influencing Patient Arrival at Hospital Emergency Department – Preliminary Study

“…These disciplines are faced with the challenge of discovering new techniques beyond the traditional ones [4]. An essential tool supporting physicians and ED staff is artifi cial intelligence [47], especially machine learning (ML) [9,13,21,22,40,43,44], which utilizes various algorithms. The past fi ve years have seen an increase in the number of applications of Artifi cial Intelligence (AI) and machine learning in various sectors of the economy, including healthcare, which has yielded many impressive advances, from autonomous driving to drug discovery.…”

Data integration using information and communication technology for emergency medical services and systems

“…Various efforts have been made to derive scientific evidence on EMS issues by integrating large-scale data sources extracted from the community [7][8][9][10] . To integrate large-scale data sources, matching through key-values is required; however, complete matching without missing values is difficult because of the heterogeneity of registries 6,11 . Even if it is possible to solve technical issues, integrating large-scale registries that include sensitive personal information (medical information) requires a latent period for administrative processes and approval from each authority.…”

“…Owing to limitations in physical space and time, it is difficult for providers to convert various events that occur in the EMS field into tabular data sources in real time. Accordingly, unstructured data that cannot be converted into a tabular form cannot be stored as a dataset, and the reliability of structured data cannot be guaranteed [ 11 ]. Therefore, the current dataset, which was manually collected in tabular form, did not fully reflect the EMS event ( Fig.…”