嵌入导联上下文编码的图卷积神经网络心律失常分类模型

doi:10.3778/j.issn.1002-8331.2309-0401

摘要/Abstract

摘要： 心律失常对患者健康造成严重威胁，其通过12导联心电图（electrocardiogram，ECG）的自动分类在临床上具有重要意义。现有研究偏重两两导联之间的相关性，忽视多导联上下文及频域特征，这导致了分析的局限性，且易受噪声干扰，影响分类准确性。该研究提出了嵌入导联上下文编码的图卷积神经网络心律失常分类模型（lead context encoding embedded graph convolutional neural network model for arrhythmia classification，LCEE-GCN）。该模型利用短时傅里叶变换获取12导联心电信号的功率谱密度（power spectral density，PSD），并运用ECG信号处理算法提取R-R间期等时域特征，通过导联上下文编码获得导联间更广泛的相关性信息，并结合PSD与时域特征构建动态图结构，利用图卷积神经网络增强模型对导联间关系的学习与表示能力。在查普曼数据集上进行的实验表明，模型达到了99.38%的准确率，超过了现有先进方法。这一创新有望提高心律失常诊断的效率和准确性。

关键词: 心律失常分类, 12导联心电图, 图卷积神经网络, 功率谱密度, 导联上下文编码

Abstract: Arrhythmia poses a serious threat to patients’health, and its automatic classification through 12-lead electrocardiograms (ECG) is of significant clinical importance. Current research has focused on the correlation between pairs of leads while overlooking the multi-lead context and frequency domain features. This has resulted in limitations in analysis and susceptibility to noise interference, affecting classification accuracy. This study introduces a graph convolutional neural network (GCN) model for arrhythmia classification that embeds lead context encoding, known as the lead context encoding embedded graph convolutional neural network model for arrhythmia classification (LCEE-GCN). Firstly, it utilizes short-time Fourier transformation to obtain the power spectral density (PSD) of the 12-lead ECG signals, and applies ECG signal processing algorithms to extract time-domain features such as R-R intervals. Then, it acquires more extensive inter-lead correlation information through lead context encoding and combines it with PSD to construct a dynamic graph structure. Finally, the model uses graph convolutional neural networks to enhance its ability to learn and represent relationships between leads. Experimental results conducted on the Chapman dataset show that the model achieves an accuracy of 99.38%, surpassing existing state-of-the-art methods. This innovation holds the potential to improve the efficiency and accuracy of arrhythmia diagnosis.

Key words: arrhythmia classification, 12-lead ECG, graph convolutional neural network (GCN), power spectral density (PSD), lead context encoding

喻云虎, 杨湘, 陈艳红. 嵌入导联上下文编码的图卷积神经网络心律失常分类模型[J]. 计算机工程与应用, 2025, 61(3): 212-222.

YU Yunhu, YANG Xiang, CHEN Yanhong. Lead Context Encoding Embedded GCN for Arrhythmia Classification[J]. Computer Engineering and Applications, 2025, 61(3): 212-222.

参考文献

[1] MURAT F, YILDIRIM O, TALO M, et al. Application of deep learning techniques for heartbeats detection using ECG signals-analysis and review[J]. Computers in Biology and Medicine, 2020, 120: 103726.
[2] 来春林. 心律失常学[M]. 北京: 科学技术文献出版社, 2011: 11-24.
LAI C L. Arrhythmology[M]. Beijing: Scientific and Technical Documentation Press, 2011: 11-24.
[3] 黄宛. 临床心电图学[M]. 北京: 人民卫生出版社, 1998.
HUANG W. Clinical electrocardiography[M].?Beijing: People’s Medical Publishing House (PRC), 1998.
[4] KOTALCZYK A. The 2020 ESC guidelines on the diagnosis and management of atrial fibrillation [J]. Arrhythmia & Electrophysiology Review, 2021, 10(2): 65.
[5] ACC/AHA/ESC guidelines for the management of patients with supraventricular arrhythmias—executive summary: a report of the American college of cardiology/American heart association task force on practice guidelines and the European society of cardiology[J]. Journal of the American College of Cardiology, 2003, 42(8): 1493-1531.
[6] CHARLE A. Effectiveness of bLCENded learning versus lectures alone on ECG analysis and interpretation by medical students[J]. BioMed Central, 2020(1).
[7] KWON J, KIM K. Artificial intelligence for early prediction of pulmonary hypertension using electrocardiography-sciencedirect[J]. The Journal of Heart and Lung Transplantation, 2020, 39(8): 805-814.
[8] 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.
[9] HANNUN Y, RAJPURKAR P, HAGHPANAHI M, et al. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network[J]. Nature Medicine, 2019, 25(1): 65-69.
[10] LIU P, SUN X, HAN Y, et al. Arrhythmia classification of LSTM autoencoder based on time series anomaly detection[J]. Biomedical Signal Processing and Control, 2022, 71: 103228.
[11] LIANG Y, ZHANG L, JIANG X, et al. Arrhythmia classification on different time windows using CSR-BiGRU network[C]//Proceedings of the 2022 15th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2022: 1-7.
[12] ZHAO Y, REN J, ZHANG B, et al. An explainable attention-based TCN heartbeats classification model for arrhythmia detection[J]. Biomedical Signal Processing and Control, 2023, 80: 104337.
[13] ANDAYESHGAR B, ABDALI-MOHAMMADI F, SEPAHVAND M, et al. Developing graph convolutional networks and mutual information for arrhythmic diagnosis based on multichannel ECG signals[J]. International Journal of Environmental Research and Public Health, 2022, 19(17): 10707.
[14] HUANG J, CHEN B, YAO B, et al. ECG arrhythmia classification using STFT-based spectrogram and convolutional neural network[J]. IEEE Access, 2019, 7: 92871-92880.
[15] KUMAR M A, CHAKRAPANI A. Classification of ECG signal using FFT based improved Alexnet classifier[J]. PLOS One, 2022, 17(9): e0274225.
[16] 曹细武, 邓亲恺. 心电图各波的频率分析[J]. 中国医学物理学杂志, 2001, 18(1): 46-48.
CAO X W, DENG Q K. Frequency analysis on the ECG waveform[J]. Chinese Journal of Medical Physics, 2001, 18(1): 46-48.
[17] LI Q, RAJAGOPALAN C, CLIFFORD G D. Ventricular fibrillation and tachycardia classification using a machine learning approach[J]. IEEE Transactions on Biomedical Engineering, 2013, 61(6): 1607-1613.
[18] HASEENA H H, MATHEW A T, PAUL J K. Fuzzy clustered probabilistic and multi layered feed forward neural networks for electrocardiogram arrhythmia classification[J]. Journal of Medical Systems, 2011, 35: 179-188.
[19] ZHANG Z, DONG J, LUO X, et al. Heartbeat classification using disease-specific feature selection[J]. Computers in Biology and Medicine, 2014, 46: 79-89.
[20] ?ZBAY Y. A new approach to detection of ECG arrhythmias: complex discrete wavelet transform based complex valued artificial neural network[J]. Journal of Medical Systems, 2009, 33(6): 435-445.
[21] GUPTA V, RATHORE N S, ARORA A K, et al. Electrocardiogram signal pattern recognition using PCA and ICA on different databases for improved health management[J]. International Journal of Applied Pattern Recognition, 2022, 7(1): 41-63.
[22] ROLAND G, PADHI S N, KAYALVILI S, et al. An automated system for arrhythmia detection using ECG records from MITDB[C]//Proceedings of the 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS), 2022: 26-33.
[23] KIRANYAZ S, INCE T, HAMILA R, et al. Convolutional neural networks for patient-specific ECG classification[C]//Proceedings of the 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2015: 2608-2611.
[24] AL RAHHAL M M, BAZI Y, AL ZUAIR M, et al. Convolutional neural networks for electrocardiogram classification[J]. Journal of Medical and Biological Engineering, 2018, 38: 1014-1025.
[25] CHEN T M, HUANG C H, SHIH E S C, et al. Detection and classification of cardiac arrhythmias by a challenge-best deep learning neural network model[J]. IScience, 2020, 23(3): 100886.
[26] LIU F, LIU C, ZHAO L, et al. An open access database for evaluating the algorithms of electrocardiogram rhythm and morphology abnormality detection[J]. Journal of Medical Imaging and Health Informatics, 2018, 8(7): 1368-1373.
[27] ZHANG J, LIANG D, LIU A, et al. MLBF-Net: a multi-lead-branch fusion network for multi-class arrhythmia classification using 12-Lead ECG[J]. IEEE Journal of Translational Engineering in Health and Medicine, 2021, 9: 1-11.
[28] ZHU J, SUN L, WANG Y, et al. A ResNet based multiscale feature extraction for classifying multi-variate medical time series[J]. KSII Transactions on Internet & Information Systems, 2022, 16(5): 1431-1445.
[29] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[30] LOUREN?O A, SILVA H, LEITE P, et al. Real time electrocardiogram segmentation for finger based ECG biometrics[C]//Proceedings of the International Conference on Bio-inspired Systems and Signal Processing, 2012: 49-54.
[31] DEFFERRARD M, BRESSON X, VANDERGHEYNST P. Convolutional neural networks on graphs with fast localized spectral filtering[C]//Advances in Neural Information Processing Systems, 2016.
[32] ZHANG J W, ZHANG J M, DANIOKO S, et al. A 12-lead electrocardiogram database for arrhythmia research covering more than 10, 000 patients[J]. Scientific Data, 2020, 7(1): 48.
[33] JANUARY C T, WANN L S, ALPERT J S, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American college of cardiology/american heart association task force on practice guidelines and the heart rhythm society[J]. Circulation, 2014, 130(23): 2071-2104.
[34] SHANKAR M G, BABU C G, RAJAGURU H. Classification of cardiac diseases from ECG signals through bio inspired classifiers with Adam and R-Adam approaches for hyperparameters updation[J]. Measurement, 2022, 194: 111048.
[35] MEQDAD M N, ABDALI-MOHAMMADI F, KADRY S. Meta structural learning algorithm with interpretable convolutional neural networks for arrhythmia detection of multisession ECG[J]. IEEE Access, 2022, 10: 61410-61425.
[36] KANG J, WEN H. A study on several critical problems on arrhythmia detection using varying-dimensional electrocardiography[J]. Physiological Measurement, 2022, 43(6): 064007.
[37] RAHUL J, SHARMA L D. Automatic cardiac arrhythmia classification based on hybrid 1-D CNN and Bi-LSTM model[J]. Biocybernetics and Biomedical Engineering, 2022, 42(1): 312-324.