Artificial intelligence algorithm for predicting cardiac arrest using electrocardiography

Abstract

Background: In-hospital cardiac arrest is a major burden in health care. Although several track-and-trigger systems are used to predict cardiac arrest, they often have unsatisfactory performances. We hypothesized that a deep-learning-based artificial intelligence algorithm (DLA) could effectively predict cardiac arrest using electrocardiography (ECG). We developed and validated a DLA for predicting cardiac arrest using ECG.

Methods: We conducted a retrospective study that included 47,505 ECGs of 25,672 adult patients admitted to two hospitals, who underwent at least one ECG from October 2016 to September 2019. The endpoint was occurrence of cardiac arrest within 24 h from ECG. Using subgroup analyses in patients who were initially classified as non-event, we confirmed the delayed occurrence of cardiac arrest and unexpected intensive care unit transfer over 14 days.

Results: We used 32,294 ECGs of 10,461 patients and 4483 ECGs of 4483 patients from a hospital were used as development and internal validation data, respectively. Additionally, 10,728 ECGs of 10,728 patients from another hospital were used as external validation data, which confirmed the robustness of the developed DLA. During internal and external validation, the areas under the receiver operating characteristic curves of the DLA in predicting cardiac arrest within 24 h were 0.913 and 0.948, respectively. The high risk group of the DLA showed a significantly higher hazard for delayed cardiac arrest (5.74% vs. 0.33%, P < 0.001) and unexpected intensive care unit transfer (4.23% vs. 0.82%, P < 0.001). A sensitivity map of the DLA displayed the ECG regions used to predict cardiac arrest, with the DLA focused most on the QRS complex.

Conclusions: Our DLA successfully predicted cardiac arrest using diverse formats of ECG. The results indicate that cardiac arrest could be screened and predicted not only with a conventional 12-lead ECG, but also with a single-lead ECG using a wearable device that employs our DLA.

Keywords: Artificial intelligence; Deep learning; Electrocardiography; Heart arrest; Hospital rapid response team.

Fig. 1

Architecture of deep-learning-based algorithm for predicting cardiac arrest. BN denotes batch normalization, Conv convolutional layer, ECG electrocardiography, and FC fully connected layer

Fig. 2

Study flowchart

Fig. 3

Performances of artificial intelligence algorithms for predicting cardiac arrest. AUC denotes area under the receiver operating characteristic curve, CI confidence interval, DLA deep-learning based artificial intelligence algorithm, NPV negative predictive value, PPV positive predictive value, and ROC receiver operating characteristic curve

Fig. 4

Cumulative hazard of deterioration event in patients who had no cardiac arrest within 24 h. DLA denotes deep-learning based artificial intelligence algorithm, ECG electrocardiography, and ICU intensive care unit. The cutoff point used for dividing the risk groups was selected when the overall sensitivity was 90% in the development dataset. Cox proportional hazards regression was used to estimate the hazard for the delayed cardiac arrest and the deterioration events

Fig. 5

Electrocardiography and sensitivity map of patient with cardiac arrest. This is electrocardiography of patient who was 62 years old and was occurred cardiac arrest in external validation hospital. The cardiac arrest occurred 18 min after acquiring electrocardiography. The deep learning based artificial intelligence algorithm predicted cardiac arrest in this patient with a value of 0.685897, which was 32.7 times the cut-off value of sensitivity 90% in development dataset