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
. 2024 Aug 1;24(15):4978.
doi: 10.3390/s24154978.

Atrial Fibrillation Prediction Based on Recurrence Plot and ResNet

Haihang Zhu  1 , Nan Jiang  1 , Shudong Xia  2 , Jijun Tong  1
Affiliations

Atrial Fibrillation Prediction Based on Recurrence Plot and ResNet

Haihang Zhu et al. Sensors (Basel). .

Abstract

Atrial fibrillation (AF) is the most prevalent form of arrhythmia, with a rising incidence and prevalence worldwide, posing significant implications for public health. In this paper, we introduce an approach that combines the Recurrence Plot (RP) technique and the ResNet architecture to predict AF. Our method involves three main steps: using wavelet filtering to remove noise interference; generating RPs through phase space reconstruction; and employing a multi-level chained residual network for AF prediction. To validate our approach, we established a comprehensive database consisting of electrocardiogram (ECG) recordings from 1008 AF patients and 48,292 Non-AF patients, with a total of 2067 and 93,129 ECGs, respectively. The experimental results demonstrated high levels of prediction precision (90.5%), recall (89.1%), F1 score (89.8%), accuracy (93.4%), and AUC (96%) on our dataset. Moreover, when tested on a publicly available AF dataset (AFPDB), our method achieved even higher prediction precision (94.8%), recall (99.4%), F1 score (97.0%), accuracy (97.0%), and AUC (99.7%). These findings suggest that our proposed method can effectively extract subtle information from ECG signals, leading to highly accurate AF predictions.

Keywords: ECG; Recurrence Plot; ResNet; atrial fibrillation; prediction.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
Flowchart for the screening and grouping process of routine 12-lead ECGs.
Figure 2
Figure 2
Schematic diagram of ECG selection for the Pre-AF group and Non-AF group (The ECGs inclusion and exclusion processes for both groups are shown, respectively. “√” represents inclusion, “×” represents exclusion).
Figure 3
Figure 3
Relationship between different types of records.
Figure 4
Figure 4
Data processing flowchart.
Figure 5
Figure 5
Wavelet decomposition schematic diagram (x-axis represents the number of data points, y-axis represents the amplitude/mv).
Figure 6
Figure 6
Comparison of the raw ECG signal and wavelet filtering.
Figure 7
Figure 7
RPs generated by different ECGs (The x-axis and y-axis represent the number of data points).
Figure 8
Figure 8
Prediction model structure diagram (the structure diagram is the proposed model-2D).
Figure 9
Figure 9
Loss values and accuracy of the model on the AFPDB dataset (overfitting is observed after 38 epochs).
Figure 10
Figure 10
Loss values and accuracy of the model on the presented dataset (overfitting is observed after 175 epochs).
Figure 11
Figure 11
(a–d) are confusion matrices of four models on AFPDB dataset; (e–h) are confusion matrices of four models on the presented dataset.
Figure 12
Figure 12
ROCs of four models on AFPDB dataset (the proposed model-2D obtains the best results, AUC = 0.997).
Figure 13
Figure 13
ROCs of four models on the presented dataset (the proposed model-2D obtains the best results, AUC = 0.96).
See this image and copyright information in PMC

References

    1. Heijman J., Luermans J.G., Linz D., van Gelder I.C., Crijns H.J. Risk factors for atrial fibrillation progression. Card. Electrophysiol. Clin. 2021;13:201–209. doi: 10.1016/j.ccep.2020.10.011. - DOI - PubMed
    1. Zhang J., Johnsen S.P., Guo Y., Lip G.Y. Epidemiology of atrial fibrillation: Geographic/Ecological risk factors, age, sex, genetics. Card. Electrophysiol. Clin. 2021;13:1–23. doi: 10.1016/j.ccep.2020.10.010. - DOI - PubMed
    1. Tsigkas G., Apostolos A., Despotopoulos S., Vasilagkos G., Kallergis E., Leventopoulos G., Mplani V., Davlouros P. Heart failure and atrial fibrillation: New concepts in pathophysiology, management, and future directions. Heart Fail. Rev. 2022;27:1201–1210. doi: 10.1007/s10741-021-10133-6. - DOI - PubMed
    1. Hu Y.-F., Chen Y.-J., Lin Y.-J., Chen S.-A. Inflammation and the pathogenesis of atrial fibrillation. Nat. Rev. Cardiol. 2015;12:230–243. doi: 10.1038/nrcardio.2015.2. - DOI - PubMed
    1. Kumar D., Puthusserypady S., Dominguez H., Sharma K., Bardram J.E. An investigation of the contextual distribution of false positives in a deep learning-based atrial fibrillation detection algorithm. Expert. Syst. Appl. 2023;211:118540. doi: 10.1016/j.eswa.2022.118540. - DOI

LinkOut - more resources