房颤是一种常见的心律失常疾病，针对现有研究工作大多依赖于单尺度信号段而忽略了不同尺度下潜在的互补信息和数据不平衡问题导致诊断性能下降的问题，提出了一种新颖的基于生成对抗网络（generative adversarial network,GAN）和多尺度残差网络（multiscaleresidual net,MS-ResNet）的房颤自动检测模型，该网络使用GAN合成具有高形态相似性的单导联心电数据来解决数据的隐私和不平衡问题。同时，设计了MS-ResNet特征提取策略，从不同尺度提取不同大小信号段的特征，从而有效地捕捉P波消失和RR间期不规则特征。该模型联合这两种策略不仅为房颤自动检测生成高质量心电图（electrocardiogram,ECG）数据，还可以利用多尺度网格提取不同波之间的时序特征。在PhysioNet Challenge2017公开ECG数据集上以及平衡后的数据集上评估了MS-ResNet的性能，并将其与现有的房颤分类模型进行了比较。实验结果表明，MS-ResNet在平衡后的数据集上平均F1值和精确率分别达到0.914 1和91.56%，与不平衡数据集相比，F1提高了4.5%，精确率提高了3.5%。
Atrial fibrillation (AF) is a common cardiac arrhythmia. However, existing research often relies on single-scale signal segments and overlooks potential complementary information at different scales as well as data imbalance issues, leading to decreased diagnostic performance. This paper proposes a novel AF automatic detection model based on generative adversarial network (GAN) and residual multi-scale network. The model utilizes GAN to synthesize single-lead ECG data with high morphological similarity, hence addressing data privacy and imbalance issues. A multi-scale residual network (MS-ResNet) feature extraction strategy was designed to extract the features of signal segments of different sizes from various scales, so as to effectively capture the features of P wave disappearance and RR interval irregularity. The model combines these two strategies not only to generate high-quality ECG (electrocardiogram) data for the automatic AF detection but also to extract temporal features between different waves using multi-scale grids. The performance of MS-ResNet is evaluated on the PhysioNet Challenge 2017 public ECG dataset and a balanced dataset, comparing it with other existing atrial fibrillation classification models. Experimental results show that the average F1 value and accuracy rate of MS-ResNet on the balanced dataset are 0.914 1 and 91.56%, respectively. Compared with the unbalanced dataset, F1 increases by 4.5%, and the accuracy rate increases by 3.5%.
[1] Ma F Y, Zhang J Y, Liang W, et al. Automated classification of atrial fibrillation using artificial neural network for wearable devices [J]. Mathematical Problems in Engineering, 2020, 12: 1-6.
[2] Ciaccio E J, Biviano, A B,Wan E Y, et al. Development of an automaton model of rotational activity driving atrial fibrillation [J]. Computers in Biology and Medicine, 2017, 83: 166-181.
[3] Kishore A, Vail A, Majid A, et al. Detection of atrial fibrillation after ischemic stroke or transient ischemic attack: a systematic review and meta-analysis [J]. Stroke, 2014, 45(2): 520-526.
[4] Buda M, Maki A, Mazurowski M A. A systematic study of the class imbalance problem in convolutional neural networks [J]. Neural Networks, 2018, 106: 249-259.
[5] Falk R H. Atrial fibrillation [J]. New England Journal of Medicine, 2001, 344(14): 1067-1078.
[6] Bahrami M, Forouzanfar M. Sleep apnea detection from single-lead ECG: a comprehensive analysis of machine learning and deep learning algorithms [J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-11.
[7] Chen S W, Wang S L, Qi X Z, et al. Review of ECG detection and classification based on deep learning: coherent taxonomy, motivation, open challenges and recommendations [J]. Biomedical Signal Processing and Control, 2022, 74: 103493.
[8] Sobahi N, Sengur A, Tan R S, et al. Attention-based 3D CNN with residual connections for efficient ECG-based COVID-19 detection [J]. Computers in Biology and Medicine, 2022, 143: 105335.
[9] Rai H M, Chatterjee K. Hybrid CNN-LSTM deep learning model and ensemble technique for automatic detection of myocardial infarction using big ECG data [J]. Applied Intelligence, 2022, 52(5): 5366-5384.
[10] Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks [DB/OL]. 2015[2023-06-29]. https:// arxiv.org/abs/1511.06434.
[11] Li W Q, Tang Y M, Yu K M, et al. SLC-GAN: an automated myocardial infarction detection model based on generative adversarial networks and convolutional neural networks with singlelead electrocardiogram synthesis [J]. Information Sciences, 2022, 589: 738-750.
[12] Rafi T H, Ko Y W. HeartNet: self multihead attention mechanism via convolutional network with adversarial data synthesis for ECG-based arrhythmia classification [J]. IEEE Access, 2022, 10: 100501-100512.
[13] Ma K, Zhan C A, Yang F. Multi-classification of arrhythmias using ResNet with CBAM on CWGAN-GP augmented ECG Gramian angular summation field [J]. Biomedical Signal Processing and Control, 2022, 77, 103684.
[14] Wang R, Yao Q, Fan X, et al. Multi-class arrhythmia detection based on neural network with multi-stage features fusion [C]//2019 IEEE International Conference on Systems, 2019: 4082-4087.
[15] Fan X M, Yao Q H, Cai Y P, et al. Multiscaled fusion of deep convolutional neural networks for screening atrial fibrillation from single lead short ECG recordings [J]. IEEE Journal of Biomedical and Health Informatics, 2018, 22(6): 1744-1753.
[16] Prabhakararao E, Dandapat S. Multiscale convolutional neural network for detecting paroxysmal atrial fibrillation from single lead ECG signals [C]//IEEE Applied Signal Processing Conference (ASPCON), 2020: 339-343.
[17] Prabhakararao E, Dandapat S. Multi-scale convolutional neural network ensemble for multiclass arrhythmia classification [J]. IEEE Journal of Biomedical and Health Informatics, 2021, 26(8): 3802-3812.
[18] Gao Y, Wang H, Liu Z. An end-to-end atrial fibrillation detection by a novel residualbased temporal attention convolutional neural network with exponential nonlinearity loss [J]. Knowledge-Based Systems, 2020, 212(1): 106589.
[19] Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting [J]. Journal of Machine Learning Research, 2014, 15(1): 1929-1958.
[20] Lourenço A, Silva H, Carreiras C, et al. Outlier detection in non-intrusive ECG biometric system [C]//International Conference Image Analysis and Recognition. Springer, 2013: 43-52.
[21] Clifford G D, Liu C Y, Moody B, et al. AF classification from a short single lead ECG recording: the physionet/computing in cardiology challenge 2017[C]//Computing in Cardiology (CinC), 2017: 1-4.
[22] Shark L K, Yu C. Denoising by optimal fuzzy thresholding in wavelet domain [J]. Electronics Letters, 2000, 36(6): 581-582.
[23] Wang R X, Fan J P, Li Y. Deep multi-scale fusion neural network for multi-class arrhythmia detection [J]. IEEE Journal of Biomedical and Health Informatics, 2020, 24(9): 2461-2472.
[24] Xie Q Y, Wang X R, Sun H Y, et al. ECG signal detection and classification of heart rhythm diseases based on ResNet and LSTM [J]. Advances in Mathematical Physics, 2021, 2021(4): 1-11.
[25] Rohr M, Reich C, Höhl A, et al. Exploring novel algorithms for atrial fibrillation detection by driving graduate level education in medical machine learning [J]. Physiological Measurement, 2022, 43(7): 074001.
[26] Wu Q, Sun Y F, Yan H, et al. ECG signal classification with binarized convolutional neural network [J]. Computers in Biology and Medicine, 2020, 121: 103800.
