A novel automated junctional ectopic tachycardia detection tool for children with congenital heart disease

Jamie L S Waugh; Raajen Patel; Yilong Ju; Ankit B Patel; Craig G Rusin; Parag N Jain

doi:10.1016/j.hroo.2022.02.014

A novel automated junctional ectopic tachycardia detection tool for children with congenital heart disease

Heart Rhythm O2. 2022 Mar 1;3(3):302-310. doi: 10.1016/j.hroo.2022.02.014. eCollection 2022 Jun.

Authors

Jamie L S Waugh¹, Raajen Patel¹, Yilong Ju², Ankit B Patel^{3

4}, Craig G Rusin⁵, Parag N Jain⁶

Affiliations

¹ Division of Research, Medical Informatics Corp, Houston, Texas.
² Department of Computer Science, Rice University, Houston, Texas.
³ Department of Neuroscience, Baylor College of Medicine, Houston, Texas.
⁴ Department of Electrical and Computer Engineering, Rice University, Houston, Texas.
⁵ Division of Cardiology, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas.
⁶ Division of Critical Care Medicine, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas.

Abstract

Background: Junctional ectopic tachycardia (JET) is a prevalent life-threatening arrhythmia in children with congenital heart disease (CHD), with marked resemblance to normal sinus rhythm (NSR) often leading to delay in diagnosis.

Objective: To develop a novel automated arrhythmia detection tool to identify JET.

Methods: A single-center retrospective cohort study of children with CHD was performed. Electrocardiographic (ECG) data produced by bedside monitors is captured automatically by the Sickbay platform. Based on the detection of R and P wave peaks, 2 interpretable ECG features are calculated: P prominence median and PR interval interquartile range (IQR). These features are used as input to a simple logistic regression classification model built to distinguish JET from NSR.

Results: This study analyzed a total of 64.5 physician-labeled hours consisting of 509,833 cardiac cycles (R-R intervals), from 40 patients with CHD. The extracted P prominence median feature is much smaller in JET compared to NSR, whereas the PR interval IQR feature is larger in JET compared to NSR. The area under the receiver operating characteristic curve for the unseen patient test cohort was 93%. Selecting a threshold of 0.73 results in a true-positive rate of 90% and a false-positive rate of 17%.

Conclusion: This novel arrhythmia detection tool identifies JET, using 2 distinctive features of JET in ECG-the loss of a normal P wave and PR relationship-allowing for early detection and timely intervention.

Keywords: Arrhythmia; Congenital heart disease; Feature extraction; Junctional ectopic tachycardia; Machine learning; Signal processing; Time series analysis.