基于增量高斯混合模型的心电分类研究
ECG Classification Based on Incremental Gaussian Mixture Model
摘要: 本研究提出一种基于增量高斯混合模型算法的心电分类方式。与其相对应的分为三个阶段：第一阶段，基于小波分解对心电信号进行预处理，以消除基线漂移(0.15~0.3 Hz)和背景噪声；第二阶段，首先基于短时修正希尔伯特变换(STMHT)定位R峰，然后自动确定QRS波进而提取心电特征参数；第三阶段提出一种无监督式增量高斯混合模型算法实现心电信号分类。本研究结果在MIT-BIH心律失常数据库中对48组心电数据进行分类比较，其准确率达93.31%。
Abstract: In this paper, an incremental Gaussian mixture model algorithm (IGMM)-based classification method is proposed for electrocardiogram (ECG) diagnosis. The three stages corresponding to the classification system are generally divided into three stages. In the first stage, wavelet decomposition-based preprocessing method is employed to eliminate the baseline drift (0.15 - 0.3 hz) and background noise. In the second stage, the automatic R peak is firstly located via the short time modified Hilbert transform (STMHT), and then the QRS complex is automatically determined to extract the ECG parameters. And an unsupervised incremental Gaussian mixture model algorithm is proposed to classify the ECG signal in the third stage. As comparative accuracies of classifying 48 groups of ECG data in MIT-BIH database, the higher classification accuracy of 93.31% greater than other methods is obtained.
文章引用：孙树平, 杜小玉, 张弼强, 陈豪, 刘叶芬, 吴越, 刘保进. 基于增量高斯混合模型的心电分类研究[J]. 建模与仿真, 2020, 9(2): 105-115. https://doi.org/10.12677/MOS.2020.92012

参考文献

[1] 胡盛寿, 杨跃进, 郑哲, 等.《中国心血管病报告2018》概要[J]. 中国循环杂志, 2019, 34(3): 209-220.
[2] Übeyli, E.D. (2007) ECG beats Classification Using Multiclass Support Vector Machines with Error Correcting Output Codes. Digital Signal Processing, 17, 675-684. [Google Scholar] [CrossRef] 
[3] DOG, B. (2010) Expert Systems with Applications ECG Beat Classification Using Particle Swarm Optimization and Radial Basis Function Neural Network. Expert Systems with Applications, 37, 7563-7569. [Google Scholar] [CrossRef] 
[4] Osowski, S., Hoai, L.T. and Markiewicz, T. (2004) Support Vector Machine-Based Expert System for Reliable Heartbeat Recognition. IEEE Transactions on Biomedical Engineering, 51, 582–589. [Google Scholar] [CrossRef] 
[5] Wu, J.F., Bao, Y.L., Chan, S.C., Wu, H.C., Zhang, L. and Wei, X.G. (2016) Myocardial Infarction Detection and Classification: A New Multi-Scale Deep Feature Learning Approach. 2016 IEEE International Conference on Digital Signal Processing (DSP), Beijing, 309-313.
[6] Poornachandra, S. (2008) Wavelet-Based Denoising Using Subband Dependent Threshold for ECG Signals. Digital Signal Processing, 18, 49-55. [Google Scholar] [CrossRef] 
[7] Engel, P.M. and Heinen, M.R. (2010) Incremental Learning of Multivariate Gaussian Mixture Models. Brazilian Symposium on Artificial Intelligence, 82-91. [Google Scholar] [CrossRef] 
[8] Bani-Hasan, A.M., El-Hefnawi, M.F. and Kadah, M.Y. (2015) Model-Based Parameter Estimation Applied on Electrocardiogram Signal.
[9] Zhidong, Z. and Min, P. (2007) ECG Denoising by Sparse Wavelet Shrinkage. 786-789.
[10] Joy, R., Acharya, U.R., Min, C., et al. (2013) Knowledge-Based Systems Characterization of ECG Beats from Cardiac Arrhythmia Using Discrete Cosine Transform in PCA Framework. Knowledge-Based Systems, 45, 76-82. [Google Scholar] [CrossRef] 
[11] Acharya, U.R., Oh, S.L., Hagiwara, Y., et al. (2017) AC SC. Computers in Biology and Medicine. Elsevier Ltd.

为你推荐