基于盲信号分离的房颤信号提取方法的研究
|
摘要
心房颤动(房颤)是指心房肌纤维出现每分钟250次到500次的不协调、不规则乱颤,是最常见的心律失常疾病之一。目前由于难以有效地跟踪房颤的状态变化和治疗效果,致使房颤的诊断准确率较低,治疗的效率和效果也都不高。
采集体表心电图(Electrocardiogram,ECG)是获取患者心脏状态信息最重要的方式之一。房颤患者的ECG主要表现为正常的窦性P波被紊乱的f波(即房颤信号)取代。f波的幅度、频谱和峰度等特征量可以从不同角度与程度上反映房颤的生理特征和机制,因此从ECG中提取f波成为房颤的计算机辅助诊疗中至关重要的一步。根据适用范围的不同,房颤信号的提取方法可分为多导联提取和单导联提取两类,适用于不同的临床情况。本文尝试将现代信号处理技术应用于房颤信号的提取中,希望能够提出综合性能更好的新的提取方法。
本文主要的研究内容如下。
提出一种可从多导联ECG中提取房颤信号的方法。该方法基于盲信号分离技术,将提取过程分为两步处理,分别采用独立成分分析和二阶盲辨识算法,不仅具有其它多导联提取方法的优点,而且提取出来的房颤信号中包含的高斯噪声成分更少。
提出一种可从单导联ECG中提取房颤信号的方法。该方法基于独立成分分析和心搏相关性,对心室波的去除更加彻底,更重要的是它消除了其它单导联提取方法的固有缺陷,即易受伪迹和异位心搏的影响。
通过实验检验这两种方法的可行性和有效性,并且与其它方法的提取效果进行对比,证明这两种方法的效果更优。
Atrial fibrillation (AF) is one of the most common sustained arrhythmias in which the atrial fibrillate randomly 250~500 times per minute. It is hard to track the natural history of arrhythmia and its response to treatment, so the diagnostic accuracy is very low, and the effect of therapy is poor and inefficient.
Acquiring electrocardiogram (ECG) is one of the most important ways to get the patient’s health condition of heart. Turbulent f wave appears in AF patient’s ECG instead of Sinus P wave. The physiological characteristic and mechanism of AF can be reflected from the amplitude, spectrum, and kurtosis of f wave. The ways to extract AF signal can be divided into two types according to different applicability: extracting AF signal from single lead ECG and from multi-lead ECG. It is intended to propose a new method of AF signal extraction in this paper, by applying modern signal processing technique.
The main contributions of this thesis are as follows:
Propose a method of extracting AF signal from multi-lead ECG. This method which is based on blind signal separation divides the processing into two steps using Independent Components Analysis and Second Order Blind Identification respectively. Besides the advantages that other multi-lead extraction methods have, there is less gaussian noise in the AF signal extracted by this method.
Propose a method of extracting AF signal from single lead ECG. This method which is based on Independent Components Analysis and heart beat correlation can cancel ventricular signal thoroughly. What’s more, it eliminates the inherent defects of other single lead extracting methods: it is easily affected by artifact and ectopic beat.
Test the feasibility and effectiveness of two methods proposed in this paper through experiment, and prove the effect of two methods is better than that of other methods.
采集体表心电图(Electrocardiogram,ECG)是获取患者心脏状态信息最重要的方式之一。房颤患者的ECG主要表现为正常的窦性P波被紊乱的f波(即房颤信号)取代。f波的幅度、频谱和峰度等特征量可以从不同角度与程度上反映房颤的生理特征和机制,因此从ECG中提取f波成为房颤的计算机辅助诊疗中至关重要的一步。根据适用范围的不同,房颤信号的提取方法可分为多导联提取和单导联提取两类,适用于不同的临床情况。本文尝试将现代信号处理技术应用于房颤信号的提取中,希望能够提出综合性能更好的新的提取方法。
本文主要的研究内容如下。
提出一种可从多导联ECG中提取房颤信号的方法。该方法基于盲信号分离技术,将提取过程分为两步处理,分别采用独立成分分析和二阶盲辨识算法,不仅具有其它多导联提取方法的优点,而且提取出来的房颤信号中包含的高斯噪声成分更少。
提出一种可从单导联ECG中提取房颤信号的方法。该方法基于独立成分分析和心搏相关性,对心室波的去除更加彻底,更重要的是它消除了其它单导联提取方法的固有缺陷,即易受伪迹和异位心搏的影响。
通过实验检验这两种方法的可行性和有效性,并且与其它方法的提取效果进行对比,证明这两种方法的效果更优。
Atrial fibrillation (AF) is one of the most common sustained arrhythmias in which the atrial fibrillate randomly 250~500 times per minute. It is hard to track the natural history of arrhythmia and its response to treatment, so the diagnostic accuracy is very low, and the effect of therapy is poor and inefficient.
Acquiring electrocardiogram (ECG) is one of the most important ways to get the patient’s health condition of heart. Turbulent f wave appears in AF patient’s ECG instead of Sinus P wave. The physiological characteristic and mechanism of AF can be reflected from the amplitude, spectrum, and kurtosis of f wave. The ways to extract AF signal can be divided into two types according to different applicability: extracting AF signal from single lead ECG and from multi-lead ECG. It is intended to propose a new method of AF signal extraction in this paper, by applying modern signal processing technique.
The main contributions of this thesis are as follows:
Propose a method of extracting AF signal from multi-lead ECG. This method which is based on blind signal separation divides the processing into two steps using Independent Components Analysis and Second Order Blind Identification respectively. Besides the advantages that other multi-lead extraction methods have, there is less gaussian noise in the AF signal extracted by this method.
Propose a method of extracting AF signal from single lead ECG. This method which is based on Independent Components Analysis and heart beat correlation can cancel ventricular signal thoroughly. What’s more, it eliminates the inherent defects of other single lead extracting methods: it is easily affected by artifact and ectopic beat.
Test the feasibility and effectiveness of two methods proposed in this paper through experiment, and prove the effect of two methods is better than that of other methods.
引文
[1]曾玉杰,魏冯宁.心房颤动抗凝治疗现状及新进展[J].中华老年心脑血管病杂志, 2009, 11(12): 922-924.
[2]陈丽丽.心房颤动发生机制和治疗方案的研究[J].国际心血管病杂志, 2010, 37(3): 152-154.
[3] Camm A J, Kirchhof P, Lip G Y, et al. Guidelines for The Management of Atrial Fibrillation: The Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology(ESC) [J]. Eur Heart, 2010, 31(19): 2369-2429.
[4]刘仁光.心房颤动伴其他心电异常的诊断[J].临床心电学杂志, 2008, 17(4): 263-265.
[5]汪剑锋.心房颤动f波振幅的研究现状[J].实用心电学杂志, 2009, 18(1): 78-79.
[6] Oosterom A Van, Lemay M, Kappenberger L. Full Spectral Analysis of the Atrial Components in the ECG During Atrial Fibrillation [C]. Computer in Cardiology 2008. Bologna: 2008: 817-820.
[7] Slocum J, Sahakian A, Swiryn S. Diagnosis of Atrial Fibrillation from Surface Electrocardiograms Based on Computer-detected Atrial Activity [J]. Electrocardiol, 1992, 25: 1-8.
[8] Thakor N V, Zhu Y S. Applications of Adaptive Filtering to ECG Analysis: Noise Cancellation and Arrhythmia Detection [J]. IEEE Tran BME, 1991, 38(8): 785-794.
[9] Stridh M, Sornmo L. Spatiotemporal QRST Cancellation Techniques for Analysis of Atrial Fibrillation [J]. IEEE Transactions on Biomedical Engineering, 2001, 48(1): 105-111.
[10] Sanchez C, Miller J, Rieta JoséJ, et al. Packet Wavelet Decomposition: An Approach for Atrial Activity Extraction [C]. IEEE Computers in Cardiology, 2002, 29: 33-36.
[11] Castells F, Rieta JoséJ, Millet J, etal. Spatiotemporal Blind Source Separation Approach to Atrial Activity Estimation in Atrial Tachyarrhythmias [J]. IEEE Transactions on Biomedical Engineering, 2005, 52 (2): 258-267.
[12] VayáCarlos, Rieta JoséJ, Sánchez César, et al. Convolutive Blind Source Separation Algorithms Applied to The Electrocardiogram of Atrial Fibrillation: Study of Performance [J]. IEEE Transactions on Biomedical Engineering, 2007, 54 (8): 1530-1533.
[13] Comon P. Independent Component Analysis: A New Concept [J]. Signal Processing, 1994, 36(3): 287.
[14] Amari S, Hyv?rinen A, Sanchez A, et al. Blind Signal Separation and Independent Component Analysis [J]. Neurocomputing, 2002, 49: 1–5.
[15] Hyv?rinen A, Karhunen J, Erkki O, Independent Component Analysis [M]. A Wiley-Interscience Publication: New York, 2001.
[16] Tong L, Soon V C, Huang Y F, etal. AMUSE: A New Blind Identification Algorithm [C]. Proc of IEEE ISCAS. New Orleans, 1990: 1784-1787.
[17]陈浩.心脏传导系统调控机制的研究[J].国际心血管病杂志, 2010, 37(3): 140-142.
[18]周宇.心外膜电位标测信号的自动联合分析方法研究[J].仪器仪表学报, 2009, 30(5): 926-931.
[19]李中键.心电图波形特征分析[J].临床心电学杂志, 2008, 24(3): 195-197.
[20]郭继鸿. EASI导联衍生12导联的原理[J].临床心电学杂志, 2008, 17(3): 163-167.
[21] Allessie M A, Karen Konings, Charles J. H, et al. Electrophysiologic Mechanisms of Perpetuation of Atrial Fibrillation [J]. American Journal Cardiology, 1996, 77: a1-a23.
[22] Chen Y H, Xu S J, Bendahhou S, et al. Gain-of-function Mutation in Familial Atrial Fibrillation [J]. Science, 2003, 299: 251-254.
[23]万红.房性心律失常信号的特征识别研究[J].临床心电学杂志, 2008, 17(3): 163-167.
[24]黄忠朝等.独立分量分析在心房纤颤中的应用研究.浙江大学学报(工学版), 2006, 46(2): 360-364.
[25] Hongjie Liu, Boqin Feng. The Research of Blind Signal Separation Algorithm Based on Neural Network [C]. ICNC 2007, 2007: 329-323.
[26] Hai Huyen Dam, Nordholm S, Siow Yong Low, et al. Blind Signal Separation Using Steepest Descent Method [J]. IEEE Transactions on Signal Processing, 2007, 55(8): 4198-4207.
[27]杨行峻,郑君里.人工神经网络与盲信号处理[M].北京:清华大学出版社, 2003.
[28]邱天爽,张旭秀,李小兵等.统计信号处理——非高斯信号处理及其应用[M].北京:电子工业出版社,2004.
[29] Thomas M, Cover J A. Elements of Information Theory [M]. A Wiley-Interscience Publication: New York, 1991.
[30]张发启.盲信号处理及应用[M].西安:西安电子科技大学出版社, 2006: 18-28.
[31]杨福生.独立分量分析的原理与应用[M].北京:清华大学出版社, 2006: 33-46.
[32] Hyv?rinen A, Pajunen P. Nonlinear Independent Component Analysis: Existence and Uniqueness Results [J]. Neural Networks, 1999, 12(3): 429–439.
[33] Zarzoso V. Comon P. Blind and Semi-blind Equalization Based on The Constant PowerCriterion [J]. IEEE Transactions on Signal Processing, 2005, 53(11): 4363–4375.
[34] Saruwatari H, Kawamura T, Nishikawa T, et al. Blind Source Separation Based on A Fast-convergence Algorithm Combining ICA and Beamforming [J]. IEEE Transactions on Audio, Speech, and Language Processing, 2006, 14(2): 1558-79169.
[35] Linsker R. An Application of The Principle of Maximum Information Preservation to Linear Systems [A]. Advances in Neural Information Processing Systems [C]. Touretzky, Morgan, 1988: 186-194.
[36] Zarzoso V, Comon P. Comparative Speed Analysis of FastICA [C]. 7th Independent Component Analysis and Signal Separation International Conference. Lundon, UK: 2007: 293-300.
[37] Hyv?rinen A. Denoising of Sensory Data by Maximum Likelihood Estimation of Sparse Components [C]. In Proc. Int. Conf. on Artificial Neural Networks, Sk?ovde, Sweden, 1998: 141–146.
[38] Hyv?rinen A. Fast and Robust Fixed-Point Algorithms for Independent Component Analysis [J]. IEEE Transactions on Neural Networks 1999, 10(3): 626-634.
[39] Wei Lu, Rajapakse J C. Approach and Applications of Constrained ICA [J]. IEEE Transactions on Neural Networks, 2005, 16(1): 203-212.
[40] Zarzoso V, Comon P, Optimal Step-size Constant Modulus Algorithm [J]. IEEE Transactions on Communications, 2008, 56(1): 10-13.
[41] Xi Lin Li, Adali T. Independent Component Analysis by Entropy Bound Minimization [J]. IEEE Transactions on Signal Processing, 2010, 58(10): 5151-5164.
[42] Langley P, Bowers E.J., Murray A. Principal Component Analysis as a Tool for Analyzing Beat-to-Beat Changes in ECG Features: Application to ECG-Derived Respiration [J]. IEEE Transactions on Biomedical Engineering, 2010, 54(7): 821-829.
[43] Lemay M. Data Processing Techniques for the Characterization of Atrial Fibrillation [D]. Lausanne: Lausanne Federal Polytechnic University, 2007.
[44]黄忠朝.基于体表心电图的心房颤动特征提取及临床辅助诊断研究[D].浙江:浙江大学, 2006.
[45] Rieta J.J., Castells, Sanchez M, et a1. Atrial Activity Extraction for Atrial Fibrillation Analysis Using Blind Source Separation [J]. IEEE Transactions on Biomedical Engineering, 2004, 51(7): 1176-1186.
[46] Langley P, Rieta J J, Stridh M, et a1. Comparison of Atrial Signal Extraction Algorithms in 12-lead ECGs with Atrial Fibrillation [J]. IEEE Trans Biomed Eng, 2006, 53(2):343-346
[47] VayáCarlos, Rieta JoséJ, Alcaraz R. Convolutive Multiband Blind Separation to Dissociate Atrial from Ventricular Activity in Atrial Fibrillation [C]. Computer in Cardiology 2009. Park City: 2009: 713-716.
[48] Alcaraz R, Rieta JoséJ, Mart?nez A. Comparative Study of Non-Invasive Organization Estimation Strategies to Predict Spontaneous Termination of Atrial Fibrillation [C]. Computer in Cardiology 2009. Park City: 2009: 513-516.
[49] Dan Xiao, Jianfeng Hu. Application of Second-Order Blind Identification in Motor Imagery EEG Classification [C]. Natural Computation, 2008. Jinan, 2008: 310-313.
[50] Corino V.D.A, Mainardi L.T, Stridh M, et al. Improved Time-Frequency Analysis of Atrial Fibrillation Signals Using Spectral Modeling [J]. IEEE Transactions on Biomedical Engineering, 2008, 55(12): 2723-2008.
[51] Howard R. Principles of Random Signal Analysis and Low Noise Design:The Power Spectral Density and its Applications [M]. A Wiley-Interscience Publication: New York, 2003: 92-137.
[52] Falendysh V, Yavorskyy B, Bachynskyy M. Digital Filters for Power Spectral Density Estimation of Heart Rhythm [C]. Telecommunications and Computer 2010, Bishkek, 2010: 280.
[53]丁志中.盲信号分离及其在心电和语音信号处理中的应用[D].合肥:中国科学技术大学, 2007.
[54]王刚.房颤信号处理及其临床应用的研究[D].成都:电子技术大学, 2008.
[55]丁志中,叶中付,汪一休.基于胸导联ECG的心房颤动信号快速盲分离[J].中国科技大学学报, 2007, 37(2): 119-125.
[56] PhysioNet, The PTB Diagnostic ECG Database [DB/OL]. (2004)[2011]. http://www.physionet.org/physiobank/database/ptbdb/.
[57] Raul Alcaraz, Rieta J J. Adaptive Singular Value Cancellation of Ventricular Activity in Single-lead Atrial Fibrillation Electrodiograms [J]. Physiological Measurement, 2008, 29: 1351-1369.
[58] Mart?nez A , Alcaraz R, Rieta JoséJ, Automatic Electrodiogram Delineator Based on Phasor Transform of Single Lead Recordings [C]. Computer in Cardiology 2010. Belfast, 2010: 987-990.
[59] Lemay M, Jean-Marc Vesin Oosterom A V, et a1. Cancel of Ventricular Activity in the ECG: Evaluation of Novel and Existing Methods [J]. IEEE Trans Biomed Eng, 2007,54(3): 542-546.
[60] Mart?nez A , Alcaraz R, Rieta JoséJ. Ectopic Beats Canceler for Improved Atrial Activity Extraction from Holter Recordings of Atrial Fibrillation [C]. Computer in Cardiology, Belfast, 2010: 1015-1018.
[61] PhysioNet, Computers in Cardiology. AF Termination Challenge Database [DB/OL]. (2004)[2011]. http://www.physionet.org/challenge/2004.
[2]陈丽丽.心房颤动发生机制和治疗方案的研究[J].国际心血管病杂志, 2010, 37(3): 152-154.
[3] Camm A J, Kirchhof P, Lip G Y, et al. Guidelines for The Management of Atrial Fibrillation: The Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology(ESC) [J]. Eur Heart, 2010, 31(19): 2369-2429.
[4]刘仁光.心房颤动伴其他心电异常的诊断[J].临床心电学杂志, 2008, 17(4): 263-265.
[5]汪剑锋.心房颤动f波振幅的研究现状[J].实用心电学杂志, 2009, 18(1): 78-79.
[6] Oosterom A Van, Lemay M, Kappenberger L. Full Spectral Analysis of the Atrial Components in the ECG During Atrial Fibrillation [C]. Computer in Cardiology 2008. Bologna: 2008: 817-820.
[7] Slocum J, Sahakian A, Swiryn S. Diagnosis of Atrial Fibrillation from Surface Electrocardiograms Based on Computer-detected Atrial Activity [J]. Electrocardiol, 1992, 25: 1-8.
[8] Thakor N V, Zhu Y S. Applications of Adaptive Filtering to ECG Analysis: Noise Cancellation and Arrhythmia Detection [J]. IEEE Tran BME, 1991, 38(8): 785-794.
[9] Stridh M, Sornmo L. Spatiotemporal QRST Cancellation Techniques for Analysis of Atrial Fibrillation [J]. IEEE Transactions on Biomedical Engineering, 2001, 48(1): 105-111.
[10] Sanchez C, Miller J, Rieta JoséJ, et al. Packet Wavelet Decomposition: An Approach for Atrial Activity Extraction [C]. IEEE Computers in Cardiology, 2002, 29: 33-36.
[11] Castells F, Rieta JoséJ, Millet J, etal. Spatiotemporal Blind Source Separation Approach to Atrial Activity Estimation in Atrial Tachyarrhythmias [J]. IEEE Transactions on Biomedical Engineering, 2005, 52 (2): 258-267.
[12] VayáCarlos, Rieta JoséJ, Sánchez César, et al. Convolutive Blind Source Separation Algorithms Applied to The Electrocardiogram of Atrial Fibrillation: Study of Performance [J]. IEEE Transactions on Biomedical Engineering, 2007, 54 (8): 1530-1533.
[13] Comon P. Independent Component Analysis: A New Concept [J]. Signal Processing, 1994, 36(3): 287.
[14] Amari S, Hyv?rinen A, Sanchez A, et al. Blind Signal Separation and Independent Component Analysis [J]. Neurocomputing, 2002, 49: 1–5.
[15] Hyv?rinen A, Karhunen J, Erkki O, Independent Component Analysis [M]. A Wiley-Interscience Publication: New York, 2001.
[16] Tong L, Soon V C, Huang Y F, etal. AMUSE: A New Blind Identification Algorithm [C]. Proc of IEEE ISCAS. New Orleans, 1990: 1784-1787.
[17]陈浩.心脏传导系统调控机制的研究[J].国际心血管病杂志, 2010, 37(3): 140-142.
[18]周宇.心外膜电位标测信号的自动联合分析方法研究[J].仪器仪表学报, 2009, 30(5): 926-931.
[19]李中键.心电图波形特征分析[J].临床心电学杂志, 2008, 24(3): 195-197.
[20]郭继鸿. EASI导联衍生12导联的原理[J].临床心电学杂志, 2008, 17(3): 163-167.
[21] Allessie M A, Karen Konings, Charles J. H, et al. Electrophysiologic Mechanisms of Perpetuation of Atrial Fibrillation [J]. American Journal Cardiology, 1996, 77: a1-a23.
[22] Chen Y H, Xu S J, Bendahhou S, et al. Gain-of-function Mutation in Familial Atrial Fibrillation [J]. Science, 2003, 299: 251-254.
[23]万红.房性心律失常信号的特征识别研究[J].临床心电学杂志, 2008, 17(3): 163-167.
[24]黄忠朝等.独立分量分析在心房纤颤中的应用研究.浙江大学学报(工学版), 2006, 46(2): 360-364.
[25] Hongjie Liu, Boqin Feng. The Research of Blind Signal Separation Algorithm Based on Neural Network [C]. ICNC 2007, 2007: 329-323.
[26] Hai Huyen Dam, Nordholm S, Siow Yong Low, et al. Blind Signal Separation Using Steepest Descent Method [J]. IEEE Transactions on Signal Processing, 2007, 55(8): 4198-4207.
[27]杨行峻,郑君里.人工神经网络与盲信号处理[M].北京:清华大学出版社, 2003.
[28]邱天爽,张旭秀,李小兵等.统计信号处理——非高斯信号处理及其应用[M].北京:电子工业出版社,2004.
[29] Thomas M, Cover J A. Elements of Information Theory [M]. A Wiley-Interscience Publication: New York, 1991.
[30]张发启.盲信号处理及应用[M].西安:西安电子科技大学出版社, 2006: 18-28.
[31]杨福生.独立分量分析的原理与应用[M].北京:清华大学出版社, 2006: 33-46.
[32] Hyv?rinen A, Pajunen P. Nonlinear Independent Component Analysis: Existence and Uniqueness Results [J]. Neural Networks, 1999, 12(3): 429–439.
[33] Zarzoso V. Comon P. Blind and Semi-blind Equalization Based on The Constant PowerCriterion [J]. IEEE Transactions on Signal Processing, 2005, 53(11): 4363–4375.
[34] Saruwatari H, Kawamura T, Nishikawa T, et al. Blind Source Separation Based on A Fast-convergence Algorithm Combining ICA and Beamforming [J]. IEEE Transactions on Audio, Speech, and Language Processing, 2006, 14(2): 1558-79169.
[35] Linsker R. An Application of The Principle of Maximum Information Preservation to Linear Systems [A]. Advances in Neural Information Processing Systems [C]. Touretzky, Morgan, 1988: 186-194.
[36] Zarzoso V, Comon P. Comparative Speed Analysis of FastICA [C]. 7th Independent Component Analysis and Signal Separation International Conference. Lundon, UK: 2007: 293-300.
[37] Hyv?rinen A. Denoising of Sensory Data by Maximum Likelihood Estimation of Sparse Components [C]. In Proc. Int. Conf. on Artificial Neural Networks, Sk?ovde, Sweden, 1998: 141–146.
[38] Hyv?rinen A. Fast and Robust Fixed-Point Algorithms for Independent Component Analysis [J]. IEEE Transactions on Neural Networks 1999, 10(3): 626-634.
[39] Wei Lu, Rajapakse J C. Approach and Applications of Constrained ICA [J]. IEEE Transactions on Neural Networks, 2005, 16(1): 203-212.
[40] Zarzoso V, Comon P, Optimal Step-size Constant Modulus Algorithm [J]. IEEE Transactions on Communications, 2008, 56(1): 10-13.
[41] Xi Lin Li, Adali T. Independent Component Analysis by Entropy Bound Minimization [J]. IEEE Transactions on Signal Processing, 2010, 58(10): 5151-5164.
[42] Langley P, Bowers E.J., Murray A. Principal Component Analysis as a Tool for Analyzing Beat-to-Beat Changes in ECG Features: Application to ECG-Derived Respiration [J]. IEEE Transactions on Biomedical Engineering, 2010, 54(7): 821-829.
[43] Lemay M. Data Processing Techniques for the Characterization of Atrial Fibrillation [D]. Lausanne: Lausanne Federal Polytechnic University, 2007.
[44]黄忠朝.基于体表心电图的心房颤动特征提取及临床辅助诊断研究[D].浙江:浙江大学, 2006.
[45] Rieta J.J., Castells, Sanchez M, et a1. Atrial Activity Extraction for Atrial Fibrillation Analysis Using Blind Source Separation [J]. IEEE Transactions on Biomedical Engineering, 2004, 51(7): 1176-1186.
[46] Langley P, Rieta J J, Stridh M, et a1. Comparison of Atrial Signal Extraction Algorithms in 12-lead ECGs with Atrial Fibrillation [J]. IEEE Trans Biomed Eng, 2006, 53(2):343-346
[47] VayáCarlos, Rieta JoséJ, Alcaraz R. Convolutive Multiband Blind Separation to Dissociate Atrial from Ventricular Activity in Atrial Fibrillation [C]. Computer in Cardiology 2009. Park City: 2009: 713-716.
[48] Alcaraz R, Rieta JoséJ, Mart?nez A. Comparative Study of Non-Invasive Organization Estimation Strategies to Predict Spontaneous Termination of Atrial Fibrillation [C]. Computer in Cardiology 2009. Park City: 2009: 513-516.
[49] Dan Xiao, Jianfeng Hu. Application of Second-Order Blind Identification in Motor Imagery EEG Classification [C]. Natural Computation, 2008. Jinan, 2008: 310-313.
[50] Corino V.D.A, Mainardi L.T, Stridh M, et al. Improved Time-Frequency Analysis of Atrial Fibrillation Signals Using Spectral Modeling [J]. IEEE Transactions on Biomedical Engineering, 2008, 55(12): 2723-2008.
[51] Howard R. Principles of Random Signal Analysis and Low Noise Design:The Power Spectral Density and its Applications [M]. A Wiley-Interscience Publication: New York, 2003: 92-137.
[52] Falendysh V, Yavorskyy B, Bachynskyy M. Digital Filters for Power Spectral Density Estimation of Heart Rhythm [C]. Telecommunications and Computer 2010, Bishkek, 2010: 280.
[53]丁志中.盲信号分离及其在心电和语音信号处理中的应用[D].合肥:中国科学技术大学, 2007.
[54]王刚.房颤信号处理及其临床应用的研究[D].成都:电子技术大学, 2008.
[55]丁志中,叶中付,汪一休.基于胸导联ECG的心房颤动信号快速盲分离[J].中国科技大学学报, 2007, 37(2): 119-125.
[56] PhysioNet, The PTB Diagnostic ECG Database [DB/OL]. (2004)[2011]. http://www.physionet.org/physiobank/database/ptbdb/.
[57] Raul Alcaraz, Rieta J J. Adaptive Singular Value Cancellation of Ventricular Activity in Single-lead Atrial Fibrillation Electrodiograms [J]. Physiological Measurement, 2008, 29: 1351-1369.
[58] Mart?nez A , Alcaraz R, Rieta JoséJ, Automatic Electrodiogram Delineator Based on Phasor Transform of Single Lead Recordings [C]. Computer in Cardiology 2010. Belfast, 2010: 987-990.
[59] Lemay M, Jean-Marc Vesin Oosterom A V, et a1. Cancel of Ventricular Activity in the ECG: Evaluation of Novel and Existing Methods [J]. IEEE Trans Biomed Eng, 2007,54(3): 542-546.
[60] Mart?nez A , Alcaraz R, Rieta JoséJ. Ectopic Beats Canceler for Improved Atrial Activity Extraction from Holter Recordings of Atrial Fibrillation [C]. Computer in Cardiology, Belfast, 2010: 1015-1018.
[61] PhysioNet, Computers in Cardiology. AF Termination Challenge Database [DB/OL]. (2004)[2011]. http://www.physionet.org/challenge/2004.