ABSTRACT
Atrial fibrillation is one of the inducing factors of many cardiac abnormalities. Currently, most atrial fibrillation classification methods only use single-lead electrocardiogram data or validate the effectiveness of the algorithm on small-scale datasets. The classification method based on single-lead atrial fibrillation ignores the complementary characteristics of different lead disease information in multi-lead electrocardiograms. If only small-scale datasets are used, the generalization ability of the algorithm cannot be guaranteed. Therefore, a multi-lead classification model for atrial fibrillation is proposed in this paper, which combines ConvNeXt, a convolutional block attention mechanism and a bidirectional long short-term memory network. Based on the characteristics and positional representation of atrial fibrillation on the electrocardiogram, a convolutional block attention mechanism is embedded in the model, enabling the model to effectively extract temporal and spatial features of the electrocardiogram during training. The specificity, sensitivity, accuracy and F1 scores obtained by this method were 98.90%, 98.04%, 98.88% and 74.32%, respectively, through testing more than 150,000 ECG recordings on the Chinese cardiovascular disease database. The experimental results indicate that this method achieves good classification performance and can provide effective technical means for the prevention and auxiliary diagnosis of atrial fibrillation.
- Islam MS, Ammour N, Alajlan N, Rhythm based heartbeat duration normalization for atrial fibrillation detection[J]. Computers in Biology and Medicine, 2016, 72: 160-169.Google ScholarDigital Library
- Moody GB, Mark RR.New method for detecting atrial fibrillation using RR intervals[J]. Computers in cardiology,1983, 10: 227-230.Google Scholar
- Lan T, Yang C. Prediction of recurrence of paroxysmal atrial fibrillation based on RR interval[J]. Journal of Biomedical Engineering, 2019, 36(4):521-530. (in Chese)Google Scholar
- Du XC, Rao NN, Qian MY, A novel method for real time atrial fibrillation detection in electrocardiograms using multiple parameters[J]. Ann Noninvasive Electrocardiology, 2014, 19(3): 217-225.Google ScholarCross Ref
- Athif M, Yasawardene PC, Daluwatte C. Detecting atrial fibrillation from short single lead ECGs using statistical and morphological features[J]. Physiological Measurement, 2018, 39(6): 064002.Google ScholarCross Ref
- Xu YF, Zhang S, Cao ZT, Extreme learning machine for heartbeat classification with hybrid time-domain and wavelet time-frequency features[J]. Journal of Healthcare Engineering, 2021, 2021(6):1-12.Google Scholar
- Zhou FY, Jin LP, Dong J. Review of convolutional neural network [J]. Chinese Journal of Computers, 2017, 40(06):1229-1251. (in Chese)Google Scholar
- Kek XY, Chin CS, Li Y. Intelligent low-complexity computing interleaving wavelet scattering based mobile shuffling network for acoustic scene classification[J]. IEEE Access, 2022,10: 82185-82201.Google ScholarCross Ref
- Chan TK, Chin CS. Multi branch convolutional macaron net for sound event detection[J]. IEEE/ACM Transactions on Audio, Speech and Language Processing, 2021, 29:2972-2985.Google ScholarDigital Library
- Chan TK, Chin CS, Li Y. Semi-supervised NMF-CNN for sound event detection[J]. IEEE Access, 2021, 9:130529-130542.Google ScholarCross Ref
- Ma C, Wei S, Chen T, Integration of results from convolutional neural network in a support vector machine for the detection of atrial fibrillation [J]. IEEE Transactions on Instrumentation and Measurement, 2020, 70: 1-10.Google Scholar
- Clifford GD, Liu CY, Moody B, AF classification from a short single lead ECG recording: The physionet /computing in cardiology challenge 2017[C]. 2017 Computing in Cardiology (CinC) , IEEE, Rennes, France, 2017:1-4.Google ScholarCross Ref
- Liu FF, Liu CY, Zhao LN, An open access database for evaluating the algorithms of ECG rhythm and morphology abnormal detection[J]. Journal of Medical Imaging and Health Informatics, 2018, 8(7) :1368-1373.Google ScholarCross Ref
- Yu Z, Chen J, Liu Y, DDCNN: A Deep Learning Model for AF Detection From a Single-Lead Short ECG Signal [J]. IEEE Journal of Biomedical and Health Informatics, 2022, 26(10): 4987-4995.Google ScholarCross Ref
- Yang P, Wang D, Kang Z, Prediction model of paroxysmal atrial fibrillation based on pattern recognition and ensemble CNN-LSTM [J]. Journal of Zhejiang University(Engineering Science), 2020, 54(5):1039-1048.Google Scholar
- Zhao R, Xu P, Liu Y, ECG-based atrial fibrillation detection based on deep convolutional residual neural network [J]. Computer Science, 2022, 49(5):186-193.Google Scholar
- Wang L, Li X, Chen J, Study on atrial fibrillation prediction algorithm combining convolutional neural network and bi-directional long short-term memory network [J]. Journal of Jiangsu Practical Electrocardiology, 2022, 31(4): 256-261.Google Scholar
- Subramanyan L, Ganesan U. A novel deep neural network for detection of Atrial Fibrillation using ECG signals[J]. Knowledge-Based Systems, 2022, 258: 109926.Google ScholarDigital Library
- Woo S, Park J, Lee JY, et al. CBAM: convolutional block attention module[J]. arXiv:1807.06521, 2018.Google Scholar
- Zhang J, Liu X, Dong J. CCDD: an enhanced standard ECG database with its management and annotation tools [J]. International Journal on Artificial Intelligence Tools, 2012, 21(5): 1240020.Google ScholarCross Ref
- Liu Z, Mao H, Wu C, A convnet for the 2020s[C]. Proceedings of the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, USA, 2022, 11976-11986.Google ScholarCross Ref
- He K, Zhang X, Ren S, Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA, 2016, 770-778.Google ScholarCross Ref
- Liu Z, Lin Y, Cao Y, Swin transformer: Hierarchical vision transformer using shifted windows[C]. Proceedings of the Proceedings of the IEEE/CVF international conference on computer vision, 2021. 10012-10022.Google Scholar
- Ba JL, Kiros JR, Hinton GE. Layer normalization[J]. arXiv preprint arXiv:160706450, 2016.Google Scholar
- Sifre L, Mallat S. Rigid-motion scattering for texture classification [J]. arXiv preprint arXiv:14031687, 2014.Google Scholar
- Imrana Y, Xiang YP, Ali L, A bidirectional LSTM deep learning approach for intrusion detection[J], Expert Systems with Applications, 2021, 185:115524.Google ScholarDigital Library
- Gulcehre C, Moczulski M, Denil M, Noisy activation functions[C]. Proceedings of the International Conference on Machine Learning, New York, USA, 2016, 48: 3059-3068.Google Scholar
- Leng Z, Tan M, Liu C, Polyloss: A polynomial expansion perspective of classification loss functions [J]. arXiv preprint arXiv:220412511, 2022.Google Scholar
- Lin TY, Goyal P, Girshick R, Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2):318-327.Google ScholarCross Ref
- Jin L, Dong J. Deep learning research on clinical electrocardiogram analysis [J]. SCIENTIA SINICA Informationis, 2015, 45(3): 398-416. . (in Chese)Google Scholar
- Hu J, Shen L, Albanie S, Squeeze-and-Excitation networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 42(8): 2011-2023.Google ScholarDigital Library
- Jin LP, Dong J. Intelligent health vessel ABC-DE: an electrocardiogram cloud computing service [J]. IEEE transactions on cloud computing, 2020, 8(3): 861-874.Google Scholar
- Jo Y-Y, Cho Y, Lee SY, Explainable artificial intelligence to detect atrial fibrillation using electrocardiogram [J]. International Journal of Cardiology, 2021, 328: 104-110.Google ScholarCross Ref
- AI Understanding Electrocardiogram | Zhongke Maidi Artificial Intelligence Technology Exploration 12 Lead Electrocardiogram Intelligent Diagnosis [EB/OL]. http://www.medicalsystem.com.cn/news/258.html, [2023].(in Chese)Google Scholar
Index Terms
- Multi-lead atrial fibrillation classification method using ConvNeXt and BiLSTM
Recommendations
Assessing the atrial electromechanical coupling during atrial focal tachycardia, flutter, and fibrillation using electromechanical wave imaging in humans
Minimally-invasive treatments of cardiac arrhythmias such as radio-frequency ablation are gradually gaining importance in clinical practice but still lack a noninvasive imaging modality which provides insight into the source or focus of an arrhythmia. ...
Detecting Atrial Fibrillation from Single-Lead ECG Using Unbalanced Multi-classification Support Vector Machine
ICMLT '19: Proceedings of the 2019 4th International Conference on Machine Learning TechnologiesAtrial fibrillation (AF) is an common arrhythmia. The incidence of AF has been increasing with the acceleration of urbanization and social aging. Therefore, the wearable ECG acquisition devices with single-lead ECG came out for early diagnosis, ...
Visualizing intracardiac atrial fibrillation electrograms using spectral analysis
Atrial fibrillation is the most common cardiac arrhythmia, and it is associated with increased risk of stroke, heart failure, and mortality. This work describes spectral analysis techniques that are being used in conjunction with visualization ...
Comments