Impact of Data Transformation: An ECG Heartbeat Classification Approach

Yongbo Liang,Ahmed Hussain,Rabab Ward,Mohamed Elgendi,Derek Abbott,Carlo Menon

doi:10.3389/fdgth.2020.610956

Abstract

Cardiovascular diseases continue to be a significant global health threat. The electrocardiogram (ECG) signal is a physiological signal that plays a major role in preventing severe and even fatal heart diseases. The purpose of this research is to explore a simple mathematical feature transformation that could be applied to ECG signal segments in order to improve the detection accuracy of heartbeats, which could facilitate automated heart disease diagnosis. Six different mathematical transformation methods were examined and analyzed using 10s-length ECG segments, which showed that a reciprocal transformation results in consistently better classification performance for normal vs. atrial fibrillation beats and normal vs. atrial premature beats, when compared to untransformed features. The second best data transformation in terms of heartbeat detection accuracy was the cubic transformation. Results showed that applying the logarithmic transformation, which is considered the go-to data transformation, was not optimal among the six data transformations. Using the optimal data transformation, the reciprocal, can lead to a 35.6% accuracy improvement. According to the overall comparison tested by different feature engineering methods, classifiers, and different dataset sizes, performance improvement also reached 4.7%. Therefore, adding a simple data transformation step, such as the reciprocal or cubic, to the extracted features can improve current automated heartbeat classification in a timely manner.

Highlights

Electrocardiographs (ECGs) have been a staple in medical practice for around a century
In Appendix Tables A, A.1–A.3 showed the performance of three classification trials (Norm vs. atrial premature beats (APBs), normal beats (Norm) vs. atrial fibrillation (AF), and Norm vs. premature ventricular contractions (PVCs)) that were achieved based on the unbalanced dataset
A positive value meant that the transformed feature improved the classification performance, and a negative value meant that the transformed feature was not helpful in classification

Summary

Introduction

Electrocardiographs (ECGs) have been a staple in medical practice for around a century. A complete heartbeat process is initiated by the sinus node—consisting of the depolarization of atriums and ventricles and the repolarization of the ventricles—in which atrial depolarization forms a P wave, ventricular depolarization forms a QRS complex wave, and the repolarization of the ventricles forms a T wave. ECGs have been used to diagnose physical heart abnormalities [1]. When beats conform to the basic structure of a QRS complex, they are called normal beats; otherwise, they may be called arrhythmic. In an arrhythmic heartbeat—such as a beat that occurs too fast, too slow, or is irregularly timed—the morphology of the ECG waves changes

Objectives

Methods

Results

Conclusion