Abstract
Atrial fibrillation (AF) is one of the most common arrhythmic complications. The diagnosis of AF usually requires long-term monitoring on the patient’s electrocardiogram (ECG) and then either having a domain expert examine the results, or extracting key features and then using a heuristic rule or data mining method to detect. Recently, researchers have attempted to use deep learning models, such as convolution neural networks (CNN) and/or Long Short-Term Memory (LSTM) neural networks to skip the feature extraction process and achieve good classification results. In this paper we propose a hybrid CNN-LSTM model which uses the short ECG signal from the PhysioNet/CinC Challenges 2017 dataset to explore and evaluate the relative performance of four data mining algorithms and three deep learning architectures, CNN, LSTM and CNN-LSTM. Our results show that all deep learning architectures except LSTM performed much better than machine learning algorithms without needing complicated feature extraction. CNN-LSTM is the best performer, achieving 97.08% accuracy, 95.52% sensitivity, 98.57% specificity, 98.46% precision, 0.99 AUC (Area under the ROC curve) value and 0.97 F1 score. With proper design of configuration, deep learning can be effective for automatic AF detection while data mining methods require domain knowledge and an extensive feature extraction and selection process to get satisfactory results.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Rho, R.W., Page, R.L.: Asymptomatic atrial fibrillation. Prog. Cardiovasc. Dis. 48(2), 79–87 (2005)
Jean-Yves, L.H., et al.: Cost of care distribution in atrial fibrillation patients: the COCAF study. Acc. Curr. J. Rev. 147(1), 121–126 (2004)
Moody, G.B., Mark, R.G.: New method for detecting atrial fibrillation using R-R intervals. Comput. Cardiol. 10, 227–230 (1983)
PhysioNet/CinC Challenges Database (2017). http://physionet.org/challenge/2017
Lecun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436 (2015)
Ladavich, S., Ghoraani, B.: Rate-independent detection of atrial fibrillation by statistical modeling of atrial activity. Biomed. Sig. Process. Control 18, 274–281 (2015)
Maji, U., Mitra, M., Pal, S.: Automatic detection of atrial fibrillation using empirical mode decomposition and statistical approach. Procedia Technol. 10(1), 45–52 (2013)
Lake, D.E., Moorman, J.R.: Accurate estimation of entropy in very short physiological time series: the problem of atrial fibrillation detection in implanted ventricular devices. Am. J. Physiol. Heart Circulatory Physiol. 300(1), H319–H325 (2011)
Nuryani, N., et al.: RR-interval variance of electrocardiogram for atrial fibrillation detection. J. Phys.: Conf. Ser. 776(1), 012105 (2016)
Afdala, A., Nuryani, N., Nugroho, A.S.: Automatic detection of atrial fibrillation using basic shannon entropy of RR interval feature. J. Phys. Conf. Ser. 795(1), 012038 (2017)
Xia, Y., et al.: Detecting atrial fibrillation by deep convolutional neural networks. Comput. Biol. Med. 93, 84–92 (2017)
Kamaleswaran, R., Mahajan, R., Akbilgic, O.: A robust deep convolutional neural network for the classification of abnormal cardiac rhythm using varying length single lead electrocardiogram. Physiol. Measur. 39(3), 035006 (2018)
Fan, X., et al.: Multi-scaled fusion of deep convolutional neural networks for screening atrial fibrillation from single lead short ECG recordings. IEEE J. Biomed. Health Inform. 1(1), 1744–1753 (2018)
Andersen, R.S., Peimankar, A., Puthusserypady, S.: A deep learning approach for real-time detection of atrial fibrillation. Expert Syst. Appl. 115, 465–473 (2018)
Sermanet, P., Chintala, S., Lecun, Y.: Convolutional neural networks applied to house numbers digit classification. In: International Conference on Pattern Recognition (2013)
Gers, F.A., Schmidhuber, J., Cummins, F.: Learning to forget: continual prediction with LSTM. Neural Comput. 12(10), 2451–2471 (2000)
Kingma, D., Ba, J.: Adam: a method for stochastic optimization. In: 3rd International Conference on Learning Representations (2015)
Ha, R., et al.: Axillary lymph node evaluation utilizing convolutional neural networks using MRI dataset. J. Digital Imaging 31(1), 1–6 (2018)
Ballinger, B., et al.: DeepHeart: semi-supervised sequence learning for cardiovascular risk prediction. In: Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18) (2018)
Acknowledgement
This project was partially supported by the National Social Science Foundation of China (No. 17BGL087). Our deepest gratitude goes to the anonymous reviewers for their careful review, comments and suggestions that have helped improve this paper.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Wu, X., Sui, Z., Chu, CH., Huang, G. (2019). Detection of Atrial Fibrillation from Short ECG Signals Using a Hybrid Deep Learning Model. In: Chen, H., Zeng, D., Yan, X., Xing, C. (eds) Smart Health. ICSH 2019. Lecture Notes in Computer Science(), vol 11924. Springer, Cham. https://doi.org/10.1007/978-3-030-34482-5_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-34482-5_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-34481-8
Online ISBN: 978-3-030-34482-5
eBook Packages: Computer ScienceComputer Science (R0)