ECG-based machine-learning algorithms for heartbeat classification

Saira Aziz,Sajid Ahmed,Mohamed-Slim Alouini

doi:10.1038/s41598-021-97118-5

Saira Aziz, Sajid Ahmed + Show 1 more

Open Access

https://doi.org/10.1038/s41598-021-97118-5

Copy DOI

Abstract

Electrocardiogram (ECG) signals represent the electrical activity of the human hearts and consist of several waveforms (P, QRS, and T). The duration and shape of each waveform and the distances between different peaks are used to diagnose heart diseases. In this work, to better analyze ECG signals, a new algorithm that exploits two-event related moving-averages (TERMA) and fractional-Fourier-transform (FrFT) algorithms is proposed. The TERMA algorithm specifies certain areas of interest to locate desired peak, while the FrFT rotates ECG signals in the time-frequency plane to manifest the locations of various peaks. The proposed algorithm’s performance outperforms state-of-the-art algorithms. Moreover, to automatically classify heart disease, estimated peaks, durations between different peaks, and other ECG signal features were used to train a machine-learning model. Most of the available studies uses the MIT-BIH database (only 48 patients). However, in this work, the recently reported Shaoxing People’s Hospital (SPH) database, which consists of more than 10,000 patients, was used to train the proposed machine-learning model, which is more realistic for classification. The cross-database training and testing with promising results is the uniqueness of our proposed machine-learning model.

Highlights

Electrocardiogram (ECG) signals represent the electrical activity of the human hearts and consist of several waveforms (P, QRS, and T)
Several algorithms have been previously reported to detect P, QRS complex, and T waves, so as to realize noise and artifact-free ECG signals, and they have been validated over MIT-BIH arrhythmia database8–13. In8, a rapid-ramp effective algorithm was proposed for the detection of R peaks, which uses the slopes between adjacent signals to determine the occurrence of the R peaks
We showed how the two-event related moving-averages (TERMA) algorithm detection performance can be improved by exploiting fractional Fourier-transform (FrFT)

Summary

Introduction

Electrocardiogram (ECG) signals represent the electrical activity of the human hearts and consist of several waveforms (P, QRS, and T). Several algorithms have been previously reported to detect P, QRS complex, and T waves, so as to realize noise and artifact-free ECG signals, and they have been validated over MIT-BIH arrhythmia database. I n9, a combination algorithm based on empiricalmode-decomposition and the Hilbert transform was proposed to detect the R peaks in ECG signals This algorithm is complicated and involves a large number of blocks for the detection of R peaks. Along with R peaks, to detect P and T peaks, Elgendi et al, proposed some algorithms based on two event-related moving averages (TERMA)[10,11,12,13] These algorithms involve different building blocks such as filtering, enhancing, block-of-interest (BOI) generation for each peak, and thresholding. I n16–20 different classifiers such as Naive Bayes, Adaboost, support vector machines (SVM) and neural networks were used in classification

Methods

Findings

Conclusion