Robust Heartbeat Classification for Wearable Single-Lead ECG via Extreme Gradient Boosting.

Huaiyu Zhu,Fan Wu,Hanshuang Xie,Yun Pan,Yisheng Zhao,Ruohong Huan

doi:10.3390/s21165290

Abstract

Wearable electrocardiogram (ECG) monitoring devices have enabled everyday ECG collection in our daily lives. However, the condition of ECG signal acquisition using wearable devices varies and wearable ECG signals could be interfered with by severe noises, resulting in great challenges of computer-aided automated ECG analysis, especially for single-lead ECG signals without spare channels as references. There remains room for improvement of the beat-level single-lead ECG diagnosis regarding accuracy and efficiency. In this paper, we propose new morphological features of heartbeats for an extreme gradient boosting-based beat-level ECG analysis method to carry out the five-class heartbeat classification according to the Association for the Advancement of Medical Instrumentation standard. The MIT-BIH Arrhythmia Database (MITDB) and a self-collected wearable single-lead ECG dataset are used for performance evaluation in the static and wearable ECG monitoring conditions, respectively. The results show that our method outperforms other state-of-the-art models with an accuracy of 99.14% on the MITDB and maintains robustness with an accuracy of 98.68% in the wearable single-lead ECG analysis.

Highlights

Arrhythmia refers to any changes of the normal electrocardiography (ECG) signals, that is, the electrical impulses causing abnormal heart rhythms, which are characterized by transience, paroxysm, and usually with no obvious symptoms [1]
This is due to the fact that the condition of ECG signal collection is commonly different and wearable ECG signals could be interfered with by severe noises, especially for those ECG signals collected using wearable single-lead ECG monitoring devices deployed in the environment of daily life usage
Automated wearable single-lead ECG signal analysis is of great significance to the monitoring of everyday cardiac activity for the detection of abnormal heart conditions, in which case

Summary

Introduction

Arrhythmia refers to any changes of the normal electrocardiography (ECG) signals, that is, the electrical impulses causing abnormal heart rhythms, which are characterized by transience, paroxysm, and usually with no obvious symptoms [1]. With the innovation of mobile health technologies, clinical-level wearable ECG monitoring devices with limited lead channels have been designed in a variety of physical forms, e.g., card-type [3], watch-type [4], and patch-type [5] and applied in dedicated clinical diagnosis and treatment scenes like immediate real-time monitoring and ultra-long-term monitoring. These devices are becoming the main source of everyday ECG signals gradually. Automated wearable single-lead ECG signal analysis is of great significance to the monitoring of everyday cardiac activity for the detection of abnormal heart conditions, in which case

Methods

Results

Discussion

Conclusion