生物医学工程学杂志

生物医学工程学杂志

基于集成学习的临床心电图分类算法研究

查看全文

随着心电图数据量快速增长,计算机辅助心电图分析也有着越来越广阔的应用需求。本文在基于导联卷积神经网络的临床心电图分类算法上提出多种策略,进一步提升其在实际应用中的性能。首先用不同的预处理方法和训练方法获得两个不同的分类器,接着用多重输出预测法来增强每个分类器的性能,最后用贝叶斯方法进行融合。测试了超过15万条心电图记录,所提方法的准确率和受试者工作特征曲线下面积(AUC)分别为85.04%和0.918 5,明显优于基于特征提取的传统方法。

With the increasing number of electrocardiogram (ECG) data, extensive application requirements of computer-aided ECG analysis have occurred. In the paper, we propose a variety of strategies to improve the performance of clinical ECG classification algorithm based on Lead Convolutional Neural Network (LCNN). Firstly, we obtained two classifiers by using different preprocessing methods and training methods in the study. Then, we applied the multiple output prediction method to both of them independently. Finally, the Bayesian approach was employed to fuse them. Tests conducted using more than 150 000 ECG records showed that the proposed method had an accuracy of 85.04% and the area under receiver operating characteristic curve (AUC) was 0.918 5, which significantly outperforms traditional methods based on feature extraction techniques.

关键词: 心电图; 集成学习; 深度学习; 卷积神经网络; 分类

Key words: electrocardiogram; ensemble learning; deep learning; Convolutional Neural Networks; classification

引用本文: 金林鹏, 董军. 基于集成学习的临床心电图分类算法研究. 生物医学工程学杂志, 2016, 33(5): 825-833. doi: 10.7507/1001-5515.20160134 复制

登录后 ,请手动点击刷新查看全文内容。 没有账号,
登录后 ,请手动点击刷新查看图表内容。 没有账号,
1. 董军, 徐淼, 詹聪明, 等.心电图识别与分类:方法、问题和新途径[J].生物医学工程学杂志, 2007, 24(6):1224-1229.
2. 王丽苹, 董军.心电图模式分类方法研究进展与分析[J].中国生物医学工程学报, 2010, 29(6):916-925.
3. CLIFFORD G D, AZUAJE F, MCSHARRY P E. Advanced methods and tools for ECG data analysis[M]. London:Artech House, 2006.
4. ZHANG Zhancheng, DONG Jun, LUO Xiaoqing, et al. Heartbeat classification using disease-specific feature selection[J]. Comput Biol Med, 2014, 46:79-89.
5. MAR T, ZAUNSEDER S, MART?NEZ J P, et al. Optimization of ECG classification by means of feature selection[J]. IEEE Trans Biomed Eng, 2011, 58(8):2168-2177.
6. MINHAS F U, ARIF M. Robust electrocardiogram (ECG) beat classification using discrete wavelet transform[J]. Physiol Meas, 2008, 29(5):555-570.
7. MARTIS R J, CHAKRABORTY C, RAY A K. A two-stage mechanism for registration and classification of ECG using Gaussian mixture model[J]. Pattern Recognit, 2009, 42(11):2979-2988.
8. KUNDU M, NASIPURI M, BASU D K. Knowledge-based ECG interpretation:a critical review[J]. Pattern Recognit, 2000, 33(3):351-373.
9. TSIPOURAS M G, FOTIADIS D I, SIDERIS D. An arrhythmia classification system based on the RR-interval signal[J]. Artif Intell Med, 2005, 33(3):237-250.
10. HUANG Huifang, LIU Jie, ZHU Qiang, et al. Detection of inter-patient left and right bundle branch block heartbeats in ECG using ensemble classifiers[J]. Biomed Eng Online, 2014, 13:1-22.
11. MARTIS R J, ACHARYA U R, ADELI H A, et al. Computer aided diagnosis of atrial arrhythmia using dimensionality reduction methods on transform domain representation[J]. Biomed Signal Process Control, 2014, 13:295-305.
12. HUANG Huifang, LIU Jie, ZHU Qiang, et al. A new hierarchical method for inter-patient heartbeat classification using random projections and RR intervals[J]. Biomed Eng Online, 2014, 13:1-26.
13. DAQROUQ K, ALKHATEEB A, AJOUR M N, et al. Neural network and wavelet average framing percentage energy for atrial fibrillation classification[J]. Comput Methods Programs Biomed, 2014, 113(3):919-926.
14. DE CHAZAL P, O'DWYER M, REILLY R B. Automatic classification of heartbeats using ECG morphology and heartbeat interval features[J]. IEEE Trans Biomed Eng, 2004, 51(7):1196-1206.
15. YE Can, KUMAR B V, COIMBRA M T. Heartbeat classification using morphological and dynamic features of ECG signals[J]. IEEE Trans Biomed Eng, 2012, 59(10):2930-2941.
16. SINGH Y N. Individual identification using linear projection of heartbeat features[J]. Applied Computational Intelligence and Soft Computing, 2014:1-14.
17. ANSI/AAMI EC57/Ed.3. Testing and reporting performance results of cardiac rhythm and ST segment measurement algorithms[S]. ANSI/AAMI, 1998, Rev 2012.
18. GOLDBERGER A L, AMARAL L A, GLASS L, et al. PhysioBank, PhysioToolkit, and PhysioNet:components of a new research resource for complex physiologic signals[J]. Circulation, 2000, 101(23):E215-E220.
19. DONG Jun, ZHANG Jiawei, ZHU Honghai, et al. Wearable ECG monitors and its remote diagnosis service platform[J]. IEEE Intell Syst, 2012, 27(6):36-43.
20. ZHANG Jiawei, LIU Xia, DONG Jun. CCDD:an enhanced Standard ECG database with its management and annotation tools[J]. Int J Artif Intell Tools, 2012, 21(5):1-26.
21. 朱洪海.心电图自动识别的关键算法设计及多体征远程监护系统研制[D].北京:中国科学院大学, 2013.
22. 王丽苹.融合领域知识的心电图分类方法研究[D].上海:华东师范大学, 2013.
23. 金林鹏, 董军.面向临床心电图分析的深层学习算法研究[J].中国科学:信息科学, 2015, 45(3):398-416.
24. THAKOR N V, WEBSTER J G, TOMPKINS W J. Estimation of QRS complex power spectra for design of a QRS filter[J]. IEEE transactions on bio-medical engineering, 1984, 31(11):702-706.
25. 国家食品药品监督管理局.YY 1139-2000单道和多道心电图机[S].北京:国家食品药品监督管理局, 2000.
26. LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[C]//Proceedings of the IEEE. New York:IEEE Press, 1998:2278-2324.
27. RUMELHART D, HINTON G, WILLIAMS R. Parallel distributed processing[M]. Cambridge:MIT Press, 1986:318-362.
28. LI M, ZHANG T, CHEN Y Q, et al. Efficient mini-batch training for stochastic optimization[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA, 2014:661-670.
29. VOGL T P, MANGIS J K, RIGLER A K, et al. Accelerating the convergence of the back-propagation method[J]. Biol Cybern, 1988, 59(4/5):257-263.
30. KITTLER J, HATEF M, DUIN R, et al. On combining classifiers[J]. IEEE Trans Pattern Anal Mach Intell, 1998, 20(3):226-239.
31. BATUWITA R, PALADE V. A new performance measure for class imbalance learning. Application to bioinformatics problem[C]//International Conference on Machine Learning and Applications. Miami Beach, 2009:545-550.
32. LIU X. Atlas of classical electrocardiograms[M]. Shanghai:Shanghai Science and Technology Press, 2011.
33. ZHU Honghai, DONG Jun. An R-peak detection method based on peaks of Shannon energy envelope[J]. Biomed Signal Process Control, 2013, 8(5):466-474.
34. 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[J]. Am J Physiol Heart Circ Physiol, 2011, 300(1):H319-H325.
35. HAKACOVA N, TRÄGÅRDH-JOHANSSON E, WAGNER G S, et al. Computer-based rhythm diagnosis and its possible influence on nonexpert electrocardiogram readers[J]. J Electrocardiol, 2012, 45(1):18-22.