Can Deep Learning Models Differentiate Atrial Fibrillation from Atrial Flutter?

• Estela Ribeiro USP
• Quenaz Bezerra Soares USP
• Felipe Meneguitti Dias USP
• Jose E. Krieger USP
• Marco Antonio Gutierrez USP

DOI: https://doi.org/10.5753/eniac.2025.11814

Abstract

Atrial Fibrillation (AFib) and Atrial Flutter (AFlut) are prevalent arrhythmias that present similar clinical features, challenging automated ECG differentiation. This study investigates the classification of AFib and AFlut using 12-lead ECGs from the CinC2021 dataset and a private dataset. We evaluated both 1D and 2D deep learning models. For 1D models, LiteVGG-11 demonstrated the highest performance, achieving an Acc of 77.91, AUROC of 87.17, and F1 score of 76.59. For 2D models, the EfficientNet-B2 outperformed other architectures, with an Acc of 75.20, AUROC of 85.50, and F1 of 71.59. Our results show that 1D models outperform 2D ones and that performance varies significantly across datasets, highlighting the difficulty in distinguishing AFib from AFlut.

References

Brundel, B. J. J. M., Ai, X., Hills, M. T., Kuipers, M. F., Lip, G. Y. H., and de Groot, N. M. S. (2022). Atrial fibrillation. Nature Reviews Disease Primers, 21(8).

Butkuvienė, M., Petrėnas, A., Sološenko, A., Martín-Yebra, A., Marozas, V., and Sörnmo, L. (2021). Considerations on performance evaluation of atrial fibrillation detectors. IEEE Transactions on Biomedical Engineering, 68(11):3250–3260.

de Chazal, P., O’Dwyer, M., and Reilly, R. (2004). Automatic classification of heartbeats using ecg morphology and heartbeat interval features. IEEE Transactions on Biomedical Engineering, 51(7):1196–1206.

Dias, F. M., Ribeiro, E., Moreno, R. A., Ribeiro, A. H., Samesima, N., Pastore, C. A., Krieger, J. E., and Gutierrez, M. A. (2023). Artificial intelligence-driven screening system for rapid image-based classification of 12-lead ecg exams: A promising solution for emergency room prioritization. IEEE Access.

Dias, F. M., Samesima, N., Ribeiro, A., Moreno, R. A., Pastore, C. A., Krieger, J. E., and Gutierrez, M. A. (2021). 2d image-based atrial fibrillation classification. In 2021 Computing in Cardiology (CinC), volume 48, pages 1–4.

Ghafoori, E., Angel, N., Dosdall, D. J., MacLeod, R. S., and Ranjan, R. (2018). Atrial fibrillation observed on surface ecg can be atrial flutter or atrial tachycardia. Journal of Electrocardiology, 51(6, Supplement):S67–S71.

Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., Mietus, J., Moody, G., Peng, C., and Stanley, H. E. (2000). Physiobank, physiotoolkit, and physionet: Components of a new research resource for complex physiologic signals. Circulation [Online], 101(23):e215–e220.

Hicks, S. A., Isaksen, J. L., Thambawita, V., Ghouse, J., Ahlberg, G., Linneberg, A., Grarup, N., Strümke, I., Ellervik, C., Olesen, M. S., Hansen, T., Graff, C., Holstein-Rathlou, N.-H., Halvorsen, P., Maleckar, M. M., Riegler, M. A., and Kanters, J. K. (2021). Explaining deep neural networks for knowledge discovery in electrocardiogram analysis. Scientific Reports, 11(1):10949.

Ivanovic, M. D., Atanasoski, V., Shvilkin, A., Hadzievski, L., and Maluckov, A. (2019). Deep learning approach for highly specific atrial fibrillation and flutter detection based on rr intervals. In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pages 1780–1783.

Jekova, I., Christov, I., and Krasteva, V. (2022). Atrioventricular synchronization for detection of atrial fibrillation and flutter in one to twelve ecg leads using a dense neural network classifier. Sensors, 22(16).

Ko Ko, N. L., Sriramoju, A., Khetarpal, B. K., and Srivathsan, K. (2022). Atypical atrial flutter: review of mechanisms, advances in mapping and ablation outcomes. Current Opinion in Cardiology, 37(1):36–45.

Moody, G. B. and Mark, R. G. (1983). A new method for detecting atrial fibrillation using r-r intervals. Computers in Cardiology, 10:227–230.

Moody, G. B. and Mark, R. G. (2000). The impact of the mit-bih arrhythmia database. IEEE Eng in Med and Biol, 20(3):45–50.

Reyna, M., Sadr, N., Perez Alday, E., Gu, A., Shah, A., Robichaux, C., Rad, A., Elola, A., Seyedi, S., Ansari, S., Ghanbari, H., Li, Q., Sharma, A., and GD, C. (2021). Will two do? varying dimensions in electrocardiography: The physionet/computing in cardiology challenge 2021. Computing in Cardiology, 48:1–4.

Reyna, M., Sadr, N., Perez Alday, E., Gu, A., Shah, A., Robichaux, C., Rad, A., Elola, A., Seyedi, S., Ansari, S., Ghanbari, H., Li, Q., Sharma, A., and GD, C. (2022). Issues in the automated classification of multilead ecgs using heterogeneous labels and populations. Physiol. Meas.

Shah, S. R., Luu, S.-W., Calestino, M., David, J., and Bray, C. (2018). Management of atrial fibrillation-flutter: uptodate guideline paper on the current evidence. Journal of Community Hospital Internal Medicine Perspectives, 8(5):269–275.

Soares, Q. B., Monteiro, R., Jatene, F. B., and Gutierrez, M. A. (2022). A lightweight unidimensional deep learning model for atrial fibrillation detection. In 2022 Computing in Cardiology (CinC), pages 1–4.

Thaler, M. S. (2019). The Only EKG Book You’ll Ever Need. Philadelphia :Wolters Kluwer Health/Lippincott Williams and Wilkins.

Tutuko, B., Nurmaini, S., Tondas, A. E., Rachmatullah, M. N., Darmawahyuni, A., Esafri, R., Firdaus, F., and Sapitri, A. I. (2021). Afibnet: an inplementation of atrial fibrillation detection with convolutional neural network. BMC Med Inform Decis Mak, 21:216.

Wang, J. and Wu, X. (2022). A deep learning refinement strategy based on efficient channel attention for atrial fibrillation and atrial flutter signals identification. Applied Soft Computing, 130:109552.

Wegner, F. K., Plagwitz, L., Doldi, C., Willy, K., Wolfes, J., Sandmann, S., Varghese, J., and Eckardt, L. (2022). Machine learning in the detection and management of atrial fibrillation. Clinical Research in Cardiology, 111(9):1010–1017.