Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2021 Apr 30:2021:6649970.
doi: 10.1155/2021/6649970. eCollection 2021.

ECG Heartbeat Classification Based on an Improved ResNet-18 Model

Enbiao Jing  1 , Haiyang Zhang  2 , ZhiGang Li  1 , Yazhi Liu  1 , Zhanlin Ji  1   3 , Ivan Ganchev  3   4   5
Affiliations
Comparative Study

ECG Heartbeat Classification Based on an Improved ResNet-18 Model

Enbiao Jing et al. Comput Math Methods Med. .

Abstract

Based on a convolutional neural network (CNN) approach, this article proposes an improved ResNet-18 model for heartbeat classification of electrocardiogram (ECG) signals through appropriate model training and parameter adjustment. Due to the unique residual structure of the model, the utilized CNN layered structure can be deepened in order to achieve better classification performance. The results of applying the proposed model to the MIT-BIH arrhythmia database demonstrate that the model achieves higher accuracy (96.50%) compared to other state-of-the-art classification models, while specifically for the ventricular ectopic heartbeat class, its sensitivity is 93.83% and the precision is 97.44%.

PubMed Disclaimer

Conflict of interest statement

The authors declare that there is no conflict of interest regarding the publication of this article.

Figures

Figure 1
Figure 1
Single heartbeat waveform.
Figure 2
Figure 2
ECG signal denoising process.
Figure 3
Figure 3
ECG signal comparison before denoising (red color) and after denoising (black color).
Figure 4
Figure 4
Heartbeat classification based on the proposed model.
Figure 5
Figure 5
The overall structure of a convolutional neural network (CNN).
Figure 6
Figure 6
The ResNet building block (adapted from [34]).
Figure 7
Figure 7
Parameters of each layer of the improved ResNet-18 model.
Figure 8
Figure 8
The structure of improved ResNet-18 model.
Figure 9
Figure 9
The improved ResNet-18 model's loss as a function of the number of iterations.
Figure 10
Figure 10
The improved ResNet-18 model's accuracy as a function of the number of iterations.
Algorithm 1
Algorithm 1
The training algorithm of the improved ResNet-18 model.
See this image and copyright information in PMC

References

    1. Behadada O., Trovati M., Kontonatsios G., Korkontzelos Y. A multinomial logistic regression approach for arrhythmia detection. International Journal of Distributed Systems and Technologies. 2017;8(4):17–33. doi: 10.4018/IJDST.2017100102. - DOI
    1. Saya S., Hennebry T. A., Lozano P., Lazzara R., Schechter E. Coronary slow flow phenomenon and risk for sudden cardiac death due to ventricular arrhythmias: a case report and review of literature. Clinical Cardiology. 2008;31(8):352–355. doi: 10.1002/clc.20266. - DOI - PMC - PubMed
    1. Sanders R. A., Kurosawa T. A., Sist M. D. Ambulatory electrocardiographic evaluation of the occurrence of arrhythmias in healthy Salukis. Journal of the American Veterinary Medical Association. 2018;252(8):966–969. doi: 10.2460/javma.252.8.966. - DOI - PubMed
    1. Sannino G., de Pietro G. A deep learning approach for ECG-based heartbeat classification for arrhythmia detection. Future Generation Computer Systems. 2018;86(Sep.):446–455. doi: 10.1016/j.future.2018.03.057. - DOI
    1. Wang J., Ye Y., Pan X., Gao X. Parallel-type fractional zero-phase filtering for ECG signal denoising. Biomedical Signal Processing and Control. 2015;18:36–41. doi: 10.1016/j.bspc.2014.10.012. - DOI

LinkOut - more resources