Automatic Detection for Multi-Labeled Cardiac Arrhythmia Based on Frame Blocking Preprocessing and Residual Networks

Zicong Li,Henggui Zhang

doi:10.3389/fcvm.2021.616585

Abstract

Introduction: Electrocardiograms (ECG) provide information about the electrical activity of the heart, which is useful for diagnosing abnormal cardiac functions such as arrhythmias. Recently, several algorithms based on advanced structures of neural networks have been proposed for auto-detecting cardiac arrhythmias, but their performance still needs to be further improved. This study aimed to develop an auto-detection algorithm, which extracts valid features from 12-lead ECG for classifying multiple types of cardiac states.Method: The proposed algorithm consists of the following components: (i) a preprocessing component that utilizes the frame blocking method to split an ECG recording into frames with a uniform length for all considered ECG recordings; and (ii) a binary classifier based on ResNet, which is combined with the attention-based bidirectional long-short term memory model.Result: The developed algorithm was trained and tested on ECG data of nine types of cardiac states, fulfilling a task of multi-label classification. It achieved an averaged F1-score and area under the curve at 0.908 and 0.974, respectively.Conclusion: The frame blocking and bidirectional long-short term memory model represented an improved algorithm compared with others in the literature for auto-detecting and classifying multi-types of cardiac abnormalities.

Highlights

Electrocardiograms (ECG) provide information about the electrical activity of the heart, which is useful for diagnosing abnormal cardiac functions such as arrhythmias
Cardiac arrhythmias refer to irregular heart rhythms, representing abnormal cardiac electrical activities associated with abnormal initiation and conduction of excitation waves in the heart [1]
The proposed frame blocking method minimizes the loss of valid signals while maintaining the continuity of ECG signals in the process of unifying the length of variant ECG recordings; (ii) we developed a neural network based on the residual networks [31] with attention-based bidirectional long short-term memory (BiLSTM)

Summary

Introduction

Electrocardiograms (ECG) provide information about the electrical activity of the heart, which is useful for diagnosing abnormal cardiac functions such as arrhythmias. Detection and risk stratification of cardiac arrhythmias are Automatic Detection of Multi-Types of Cardiac Arrhythmia crucial for averting severe cardiac consequences With their ability to represent useful information regarding the electrical activity of the heart, electrocardiograms (ECG) measured via electrodes placed on the body surface played an important role in diagnosing cardiac abnormalities [3]. To extract sufficient features automatically and achieve high classification accuracy, recent advancements in deep neural network [8] helped to develop several improved auto-detection algorithms [5, 9, 10] for ECG analysis and classification These studies illustrated that the deeplearning-based algorithms have the advantages of extracting and processing ECG features automatically

Objectives

Findings

Discussion

Conclusion