June 18, 2019Open Access

A Novel Feature Representation for Single-Channel Heartbeat Classification based on Adaptive Fourier Decomposition

Q: What is the clinical evidence from this study?

Study design: Other. Population: Arrhythmia (n=100103). Intervention: Adaptive Fourier Decomposition (AFD) based feature representation with SVM classifier vs. State-of-the-art automatic heartbeat classification algorithms. Primary outcome: Overall accuracy of heartbeat classification.

CTChunyu TanAnhui University LZLiming ZhangUniversity of Macau HWHau‐Tieng WuGeneral Cardiology

Key Result

The proposed adaptive Fourier decomposition-based feature representation with a support vector machine classifier achieved an overall accuracy of 85.02% for single-lead ECG heartbeat classification.

Structured PICO

Does an SVM classifier trained with AFD-derived IF features accurately classify heartbeats from single-lead ECG?

Population

100,103 heartbeats from the MIT-BIH arrhythmia database and 12,148 class S heartbeats from the MIT-BIH supraventricular database

Intervention

Support vector machine (SVM) classifier trained with Adaptive Fourier Decomposition (AFD)-derived instantaneous frequency (IF) features, ECG landmark features, and RR interval features

Comparator

State-of-the-art automatic heartbeat classification algorithms based on two-leads ECG

Outcome

Heartbeat classification performance (sensitivities and positive predictive values) according to the Association for the Advancement of Medical Instrumentation (AAMI) standardsurrogate

A novel heartbeat classification algorithm using Adaptive Fourier Decomposition on single-lead ECG achieves performance comparable to state-of-the-art two-lead ECG algorithms.

Limitations

The data is from a publicly available database that is not collected from equipment tailored for telemedicine.
The database size is limited.
The study focuses only on the ECG signal.

Abstract

This paper proposes a novel approach for heartbeat classification from single-lead electrocardiogram (ECG) signals based on the novel adaptive Fourier decomposition (AFD). AFD is a recently developed signal processing tool that provides useful morphological features, referred to as AFD-derived instantaneous frequency (IF) features, that are different from those provided by traditional tools. A support vector machine (SVM) classifier is trained with the AFD-derived IF features, ECG landmark features, and RR interval features. To evaluate the performance of the trained classifier, the Association for the Advancement of Medical Instrumentation (AAMI) standard is applied to the publicly available benchmark databases, including MIT-BIH arrhythmia database and MIT-BIH supraventricular arrhythmia database, to classify heartbeats from single-lead ECG. The overall performance in terms of sensitivities and positive predictive values is comparable to the state-of-the-art automatic heartbeat classification algorithms based on two-leads ECG.

Ask AI

Helpful

Bookmark

View Full Paper

Ask AI

Helpful

Bookmark

View Full Paper

Cite This Study

Tan et al. (Tue,) conducted a other in Arrhythmia (n=100,103). Adaptive Fourier Decomposition (AFD) based feature representation with SVM classifier vs. State-of-the-art automatic heartbeat classification algorithms was evaluated on Overall accuracy of heartbeat classification. The proposed adaptive Fourier decomposition-based feature representation with a support vector machine classifier achieved an overall accuracy of 85.02% for single-lead ECG heartbeat classification.

synapsesocial.com/papers/6a1c687066d062ff2dc3c603 https://doi.org/https://doi.org/10.48550/arxiv.1906.07361

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: