独立分量分析在心电和外阴诱发电位信号处理中的应用
|
摘要
独立分量分析(ICA)是近年来由盲源分离(BSS)技术发展来的一种多维信号处理方法。它以非高斯信号为处理对象,在满足一定的条件下,能从多路观测信号中,较完整的分离出隐含其中的若干独立源信号。
二十世纪八十年代起,ICA从提出基本概念、研究算法理论,到如今走向应用领域,吸引了越来越多的学者和研究团体,他们从不同的角度对ICA的基本理论、算法、应用进行了深入研究。本文在阐述ICA理论基础的同时,以心电和外阴诱发电位信号的处理为实例,研究了ICA方法在生物医学信号处理中的应用。
生物医学信号的产生过程中,包含了人体其他器官组织产生的生物电信号以及各种外界因素引起的干扰信号。我们对其处理的目的是为了从复杂的背景噪声中分离出有用的原始信号,进而从中提取出具有明确生理意义的信号特征,并应用于临床医学的研究。
本文所做的主要研究工作如下:
1、从基本原理出发,探讨了ICA算法的目标函数和优化算法,阐述了FastICA等算法的基本原理,并对FastICA算法的盲源分离性能进行了仿真验证,结果证明了ICA的有效性。
2、研究了ICA在心肌梗塞病人的心电信号分离中的应用,得出了ICA分析方法对于消除心电信号噪声有较好效果的结论;同时将心电信号看成是心电传导通路上不同部位兴奋产生的心电向量的投影,单独研究分离出来的子信号成份有利于获得更加丰富的信息,这可以给临床提供更加灵活多样并且特别的诊断手段。3、研究了ICA在诱发电位信号提取中的应用,提出了将ICA方法应用到外阴诱发电位信号提取中去的方法,为用ICA方法取代叠加平均方法,降低刺激次数,提高受检者的依从性,改善信号提取质量以及该项技术临床应用提供理论基础。
Independent component analysis (ICA) is a multi-dimensional signal- processing method which was developed from the Blind Source Separation (BSS) technology in recent years . Under certain conditions , it can separate the multi-channel non-Gaussian observated signals to some hidden independent sources signals .
From the 1980s , people has being proposing the basic concept about ICA , researching algorithm theory on ICA , and now ICA comes to the application field . It attracts a growing number of scholars and research groups who study on the ICA's basic theory , algorithm , application from different angles . In this paper , we expounded basic theory about ICA , and studied it’s applications in biomedical signal processing with Electrocardiography and Pudendal Evoked Potential Signal as examples.
During the process of generating Biomedical signals , it mixed biological signals from other human organs and tissues as well as the interference signal from various external factors . The aim of our dealing with it is to isolate the useful original signal from the complex background noise , and extract a clear physiological significance of the signal characteristics , and use it in clinical studies.
In this paper , the main research work is as follows:
1、Based on the principle , we discussed the goal functions and the optimization algorithm of the ICA , described the basic principles of the algorithm such as FastICA , simulated it and demonstrated the effectiveness of the ICA.
2、we studied the ICA’s applications in the ECG signals of patients with myocardial infarction , concluded that the ICA method has a good effects in the elimination of ECG noise . At the same time , we regarded the ECG signal as the Projection of the ECG signal vectors raised from the different excited parts on the transduction pathway . The study of the separated signal components is useful to obtain more abundant information , which can provide a more flexible , diverse and special diagnostic tools for clinical study.
3、We studied the ICA’s applications in the evoked potential signal extraction , proposed the method of the application of extracting Pudendal Evoked Potential Signals , and providing the theoretical basis for alternating the average method with ICA , reducing the frequency of stimulation, increasing the compliance of subjects , improving the quality of the extracted signal , as well as applying for the clinical study.
二十世纪八十年代起,ICA从提出基本概念、研究算法理论,到如今走向应用领域,吸引了越来越多的学者和研究团体,他们从不同的角度对ICA的基本理论、算法、应用进行了深入研究。本文在阐述ICA理论基础的同时,以心电和外阴诱发电位信号的处理为实例,研究了ICA方法在生物医学信号处理中的应用。
生物医学信号的产生过程中,包含了人体其他器官组织产生的生物电信号以及各种外界因素引起的干扰信号。我们对其处理的目的是为了从复杂的背景噪声中分离出有用的原始信号,进而从中提取出具有明确生理意义的信号特征,并应用于临床医学的研究。
本文所做的主要研究工作如下:
1、从基本原理出发,探讨了ICA算法的目标函数和优化算法,阐述了FastICA等算法的基本原理,并对FastICA算法的盲源分离性能进行了仿真验证,结果证明了ICA的有效性。
2、研究了ICA在心肌梗塞病人的心电信号分离中的应用,得出了ICA分析方法对于消除心电信号噪声有较好效果的结论;同时将心电信号看成是心电传导通路上不同部位兴奋产生的心电向量的投影,单独研究分离出来的子信号成份有利于获得更加丰富的信息,这可以给临床提供更加灵活多样并且特别的诊断手段。3、研究了ICA在诱发电位信号提取中的应用,提出了将ICA方法应用到外阴诱发电位信号提取中去的方法,为用ICA方法取代叠加平均方法,降低刺激次数,提高受检者的依从性,改善信号提取质量以及该项技术临床应用提供理论基础。
Independent component analysis (ICA) is a multi-dimensional signal- processing method which was developed from the Blind Source Separation (BSS) technology in recent years . Under certain conditions , it can separate the multi-channel non-Gaussian observated signals to some hidden independent sources signals .
From the 1980s , people has being proposing the basic concept about ICA , researching algorithm theory on ICA , and now ICA comes to the application field . It attracts a growing number of scholars and research groups who study on the ICA's basic theory , algorithm , application from different angles . In this paper , we expounded basic theory about ICA , and studied it’s applications in biomedical signal processing with Electrocardiography and Pudendal Evoked Potential Signal as examples.
During the process of generating Biomedical signals , it mixed biological signals from other human organs and tissues as well as the interference signal from various external factors . The aim of our dealing with it is to isolate the useful original signal from the complex background noise , and extract a clear physiological significance of the signal characteristics , and use it in clinical studies.
In this paper , the main research work is as follows:
1、Based on the principle , we discussed the goal functions and the optimization algorithm of the ICA , described the basic principles of the algorithm such as FastICA , simulated it and demonstrated the effectiveness of the ICA.
2、we studied the ICA’s applications in the ECG signals of patients with myocardial infarction , concluded that the ICA method has a good effects in the elimination of ECG noise . At the same time , we regarded the ECG signal as the Projection of the ECG signal vectors raised from the different excited parts on the transduction pathway . The study of the separated signal components is useful to obtain more abundant information , which can provide a more flexible , diverse and special diagnostic tools for clinical study.
3、We studied the ICA’s applications in the evoked potential signal extraction , proposed the method of the application of extracting Pudendal Evoked Potential Signals , and providing the theoretical basis for alternating the average method with ICA , reducing the frequency of stimulation, increasing the compliance of subjects , improving the quality of the extracted signal , as well as applying for the clinical study.
引文
[1] C.Jutten and J.Herault,Blind separation of sources,Signal Processing,1991,24:1~10
[2] P.Comon,Independent component analysis,A new concept,Signal Processing,1994,36:287~314
[3] A.J.Bell and T.J.Sejnowski, An information-maximization approach to blind separation and blind deconvolution,Neural Computation,1995,7(6):1004~1034
[4] J.Cardoso and B.Hvam.Laheld,Equivariant adaptive source separation,IEEE Trans.On Signal Processing,1996,44(12):3017~3030
[5] S.Amari , Natural gradient works efficiently in learning , Neural Computation,1998,10:251~276
[6] J.Cardoso,Blind signal separation: Statistical principles,Proc. of the IEEE,1998,86(10):2009~2025
[7] S.Amari and A.Cichocki,Adaptive blind signal processing-neural network approaches,Proc. of the IEEE,1998,86(10):2026~2048
[8] Hyvarinen,J.Karhunen and E.Oja,Independent Component Analysis,John Wiley & Sons.inc,2001
[9] Hyvarinen and E.Oja , A fast fixed-point algorithm for independent component analysis,Neural Computation,1997,9(7):1483~1492
[10] J.F.Cardoso , Blind beamforming for non-Gaussian signals , IEEE Proceedings-F,1993,140(6):224~230
[11] Belouchrani,K.Abed-Meraim,J.-F.Cardoso,et al,A blind source separation technique using second-order statistics,IEEE Trans.On Signal Processing,1997,45(2):434~444
[12]高晋占,微弱信号检测,北京,清华大学出版社,2004
[13]徐泾平,吴延军,生物医学信号的检测、处理与应用,国外医学生物医学工程分册,1995,18:68～75
[14]张佳华,张仲乐,采用加权平均算法提取事件相关电位,医疗卫生装备,2003,11:5～6
[15]王荣昌,都思丹,基于参数模型和独立分量分析的事件相关诱发电位单次提取,生物医学工程学杂志,2006,23(6):1222～1227
[16]吴祈耀,吴祈宗,包含离散谱线的电生理信号的参数模型谱估计问题,北京生物医学工程,1994,13(2):70～73
[17]吴芝芳,孔兰,郑旭媛等,提取脑电信号的新方法研究─—自适应卡尔曼滤波,信号处理,1994,10(4):238～243
[18]张佳华杨仲乐,基于小波变换的单次诱发电位信号时频分析,生物物理学报,2004,20(3):203～208
[19]徐宁寿张建华曹正才等,小波变换在视觉诱发脑电信号提取中的应用──提取视觉诱发脑电信号的新方法之三,北京工业大学学报,2000,26(4):9～14
[20]张烈平张俞伟莫玮,RBF神经网络在诱发脑电信号分类中的应用研究,广东工业大学学报,2004,21(4):16～20
[21] Gharieb RR,Cichocki A,Noise reduction in brain evoked potentials based on third-order correlations , IEEE transaction on biomedical engineering,2001,48(5):501
[22] T.M.Cover,J.A.Thomas.Elements of information theory,New York:John Wiley & Sons,LTD, 1991
[23]张贤达,现代信号处理(第二版),北京:清华大学出版社,2002年,263
[24] Hyvarinen,Survey on independent component analysis,Neural Computing Surveys,1999,2:94~128
[25]张旭秀,邱天爽,张润烜等,独立分量分析原理及其应用,大连铁道学院学报,2003,24(2):6469
[26] A.K.Barros , A.Mansour,N.Ohnishi , Removing artifacts from electrocardiographic signals using independent components analysis ,Neurocomputing,1998,22:173~186
[27] Zhilin Zhang, Zhang Yi. An efficient independent component analysis algorithm for sub-Gaussian sources. Lecture Notes in Computer Science, 2005, 3496: 967~972
[28] Keshia N. Leach,A survey Paper on Independent Component Analysis[J],IEEE,2002:239~242
[29] Cichocki, R. Thawonmas,S. Amari, Sequential blind signal extraction in order specified by stochastic properties,Electronics Letters,1997,33(1):64~65
[30] J.F.Cardoso,Infomax and maximum likelihood for blind sourceseparation,IEEE signal processing letters,April,1997,4(4):112~114
[31]杨福生,洪波,独立分量分析及其在生物医学工程中的应用,国外医学生物医学工程分册,2000,23(3):129~134
[32] T.P. Jung,Colin Humphries,T.W.Lee,Removing electroencephalographic artifacts:comparison between ICA and PCA,1998:63~72
[33] S. Amari, Natural gradient works efficiently in learning, Neural Computation,1998,10:271~276
[34] T.W.Lee,M.girolami and t.j.Sejnowski,Independent component analysis using an extended infomax algorithm for mixed subgaussian an supergaussian sources,Neural Computation,1999,11:417~419
[35] Hyvarinen,Fast and robust fixed-point algorithms for independent component analysis,IEEE Trans.On Neural Networks,1999,10(3):626~634
[36]王永国略谈心电图的干扰与抗干扰, Journal of Prctical Electroc Ardiogy J S,1996,4(4):104～107
[37]曹玉珍,陈成,胡勇等,基于独立成份分析的肌电信号消噪方法研究,信号处理,2003,19(4):369～372
[38] Taigang He,G.Clifford,L.Tarassenko,Application of ICA in removing artifacts from the ECG,Neural Computing and Applications,2002,12
[39] Barros,A.Mansour,N.Onishi,Removing artifacts from electrocardiographic signals using independent component analysis,Neurocomputing,1998,22:173~189
[40] J.J.Rieta,F.Castells,C.Sanchez,V.Zarzoso and J.Millet,Atrial activity extraction for atrial fibrillation analysis using blind source separation,IEEE Tran.Biomed.Eng,2004,51:1176~1186
[41] Y. Ichimaru, G.B. Moody. Development of the polysomnographic database on CD-ROM. Psychiatry and Clinical Neurosciences 53:175-177 (April 1999)
[42] Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG, Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals. Circulation 101(23):e215-e220; 2000 (June 13)
[43]张明岛陈兴时等,脑诱发电位学,上海,上海科技教育出版社,1995
[44] Pan YF,Evoked potentials in medical practice (2th Edition),Beijing, People’s Health Publishing House,2000[潘映辐,临床诱发电位学(第二版),北京:人民卫生出版社,2000]
[45]任铭奎,代丽丽,战杰,应用视觉诱发电位检测早期视神经挫伤,中国民康医学,2007,6(19):417～418
[46]李柏坚,视觉诱发电位在弱视诊断中的应用,中国医疗前沿,2007,2(18):104～105
[47]陈立荣,胡德强,冯梅梅,脑电图与视觉诱发电位对抽动秽语综合征患儿注意力焕散现象的评估,中国临床康复,9(24):96～97
[48] Zhou Zhenhe,Zhu Hongmei,Li Hengfen,Character of auditory evoked potential P300 of patients with senile psychotic depression,Chinese Journal of Clinical Rehabilitation,2005,9(36):145~146
[49]周振和,朱红梅,李恒芬等,老年期抑郁症患者认知功能损害与听觉诱发电位P300的特征变化,中国临床康复,2005,9(20):12～14
[50]朱玉珍,视觉及脑干听觉诱发电位在运动员机能评价中的应用, [硕士学位论文],山东,山东师范大学,2007。
[51]杜明辉陈和晏胡勇,快速检测SEP在手术中脊髓监护的应用,生物医学工程学杂志,2001,18(2):169～172
[52]沈翠茹,王一林,段淑荣,体感诱发电位对糖尿病神经病变的早期诊断价值,哈尔滨医科大学学报,2001,35(1):58
[53]王玉良,陆洪英,王益光等,体感诱发电位早期诊断糖尿病神经病变的研究,数理医药学杂志,2001,14(1):18～19
[54]胡晓晴唐娜,体感诱发电位的基本原理,国外医学·物理医学与康复学分册,2005,25(2):53～55
[55] Di Lazzaro,Role of motor evoked potential in diagnoisis of cauda equina and lumbosacral cord lesions,Neurology,2004,63(12):2266~2271
[56]赵强,王明翠,孙霰等,脑诱发电位对酒依赖患者脑干功能及皮质认知功能的评估作用,中国临床康复,2004,8(25):5246～5247
[57]王飞翔张玲莉,勃起功能障碍的诊断方法及其进展,法医学杂志,2005,21(3):216～218
[58]石伟王顺德汪惠书等,神经电生理检测对勃起功能障碍的诊断价值,临床神经电生理学杂志,2004,13(3):168
[59]邵蓓李澄棣王汛,心因性阳痿患者的神经电生理表现,临床神经电生理学杂志,2002,11(3):150～151
[2] P.Comon,Independent component analysis,A new concept,Signal Processing,1994,36:287~314
[3] A.J.Bell and T.J.Sejnowski, An information-maximization approach to blind separation and blind deconvolution,Neural Computation,1995,7(6):1004~1034
[4] J.Cardoso and B.Hvam.Laheld,Equivariant adaptive source separation,IEEE Trans.On Signal Processing,1996,44(12):3017~3030
[5] S.Amari , Natural gradient works efficiently in learning , Neural Computation,1998,10:251~276
[6] J.Cardoso,Blind signal separation: Statistical principles,Proc. of the IEEE,1998,86(10):2009~2025
[7] S.Amari and A.Cichocki,Adaptive blind signal processing-neural network approaches,Proc. of the IEEE,1998,86(10):2026~2048
[8] Hyvarinen,J.Karhunen and E.Oja,Independent Component Analysis,John Wiley & Sons.inc,2001
[9] Hyvarinen and E.Oja , A fast fixed-point algorithm for independent component analysis,Neural Computation,1997,9(7):1483~1492
[10] J.F.Cardoso , Blind beamforming for non-Gaussian signals , IEEE Proceedings-F,1993,140(6):224~230
[11] Belouchrani,K.Abed-Meraim,J.-F.Cardoso,et al,A blind source separation technique using second-order statistics,IEEE Trans.On Signal Processing,1997,45(2):434~444
[12]高晋占,微弱信号检测,北京,清华大学出版社,2004
[13]徐泾平,吴延军,生物医学信号的检测、处理与应用,国外医学生物医学工程分册,1995,18:68～75
[14]张佳华,张仲乐,采用加权平均算法提取事件相关电位,医疗卫生装备,2003,11:5～6
[15]王荣昌,都思丹,基于参数模型和独立分量分析的事件相关诱发电位单次提取,生物医学工程学杂志,2006,23(6):1222～1227
[16]吴祈耀,吴祈宗,包含离散谱线的电生理信号的参数模型谱估计问题,北京生物医学工程,1994,13(2):70～73
[17]吴芝芳,孔兰,郑旭媛等,提取脑电信号的新方法研究─—自适应卡尔曼滤波,信号处理,1994,10(4):238～243
[18]张佳华杨仲乐,基于小波变换的单次诱发电位信号时频分析,生物物理学报,2004,20(3):203～208
[19]徐宁寿张建华曹正才等,小波变换在视觉诱发脑电信号提取中的应用──提取视觉诱发脑电信号的新方法之三,北京工业大学学报,2000,26(4):9～14
[20]张烈平张俞伟莫玮,RBF神经网络在诱发脑电信号分类中的应用研究,广东工业大学学报,2004,21(4):16～20
[21] Gharieb RR,Cichocki A,Noise reduction in brain evoked potentials based on third-order correlations , IEEE transaction on biomedical engineering,2001,48(5):501
[22] T.M.Cover,J.A.Thomas.Elements of information theory,New York:John Wiley & Sons,LTD, 1991
[23]张贤达,现代信号处理(第二版),北京:清华大学出版社,2002年,263
[24] Hyvarinen,Survey on independent component analysis,Neural Computing Surveys,1999,2:94~128
[25]张旭秀,邱天爽,张润烜等,独立分量分析原理及其应用,大连铁道学院学报,2003,24(2):6469
[26] A.K.Barros , A.Mansour,N.Ohnishi , Removing artifacts from electrocardiographic signals using independent components analysis ,Neurocomputing,1998,22:173~186
[27] Zhilin Zhang, Zhang Yi. An efficient independent component analysis algorithm for sub-Gaussian sources. Lecture Notes in Computer Science, 2005, 3496: 967~972
[28] Keshia N. Leach,A survey Paper on Independent Component Analysis[J],IEEE,2002:239~242
[29] Cichocki, R. Thawonmas,S. Amari, Sequential blind signal extraction in order specified by stochastic properties,Electronics Letters,1997,33(1):64~65
[30] J.F.Cardoso,Infomax and maximum likelihood for blind sourceseparation,IEEE signal processing letters,April,1997,4(4):112~114
[31]杨福生,洪波,独立分量分析及其在生物医学工程中的应用,国外医学生物医学工程分册,2000,23(3):129~134
[32] T.P. Jung,Colin Humphries,T.W.Lee,Removing electroencephalographic artifacts:comparison between ICA and PCA,1998:63~72
[33] S. Amari, Natural gradient works efficiently in learning, Neural Computation,1998,10:271~276
[34] T.W.Lee,M.girolami and t.j.Sejnowski,Independent component analysis using an extended infomax algorithm for mixed subgaussian an supergaussian sources,Neural Computation,1999,11:417~419
[35] Hyvarinen,Fast and robust fixed-point algorithms for independent component analysis,IEEE Trans.On Neural Networks,1999,10(3):626~634
[36]王永国略谈心电图的干扰与抗干扰, Journal of Prctical Electroc Ardiogy J S,1996,4(4):104～107
[37]曹玉珍,陈成,胡勇等,基于独立成份分析的肌电信号消噪方法研究,信号处理,2003,19(4):369～372
[38] Taigang He,G.Clifford,L.Tarassenko,Application of ICA in removing artifacts from the ECG,Neural Computing and Applications,2002,12
[39] Barros,A.Mansour,N.Onishi,Removing artifacts from electrocardiographic signals using independent component analysis,Neurocomputing,1998,22:173~189
[40] J.J.Rieta,F.Castells,C.Sanchez,V.Zarzoso and J.Millet,Atrial activity extraction for atrial fibrillation analysis using blind source separation,IEEE Tran.Biomed.Eng,2004,51:1176~1186
[41] Y. Ichimaru, G.B. Moody. Development of the polysomnographic database on CD-ROM. Psychiatry and Clinical Neurosciences 53:175-177 (April 1999)
[42] Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG, Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals. Circulation 101(23):e215-e220; 2000 (June 13)
[43]张明岛陈兴时等,脑诱发电位学,上海,上海科技教育出版社,1995
[44] Pan YF,Evoked potentials in medical practice (2th Edition),Beijing, People’s Health Publishing House,2000[潘映辐,临床诱发电位学(第二版),北京:人民卫生出版社,2000]
[45]任铭奎,代丽丽,战杰,应用视觉诱发电位检测早期视神经挫伤,中国民康医学,2007,6(19):417～418
[46]李柏坚,视觉诱发电位在弱视诊断中的应用,中国医疗前沿,2007,2(18):104～105
[47]陈立荣,胡德强,冯梅梅,脑电图与视觉诱发电位对抽动秽语综合征患儿注意力焕散现象的评估,中国临床康复,9(24):96～97
[48] Zhou Zhenhe,Zhu Hongmei,Li Hengfen,Character of auditory evoked potential P300 of patients with senile psychotic depression,Chinese Journal of Clinical Rehabilitation,2005,9(36):145~146
[49]周振和,朱红梅,李恒芬等,老年期抑郁症患者认知功能损害与听觉诱发电位P300的特征变化,中国临床康复,2005,9(20):12～14
[50]朱玉珍,视觉及脑干听觉诱发电位在运动员机能评价中的应用, [硕士学位论文],山东,山东师范大学,2007。
[51]杜明辉陈和晏胡勇,快速检测SEP在手术中脊髓监护的应用,生物医学工程学杂志,2001,18(2):169～172
[52]沈翠茹,王一林,段淑荣,体感诱发电位对糖尿病神经病变的早期诊断价值,哈尔滨医科大学学报,2001,35(1):58
[53]王玉良,陆洪英,王益光等,体感诱发电位早期诊断糖尿病神经病变的研究,数理医药学杂志,2001,14(1):18～19
[54]胡晓晴唐娜,体感诱发电位的基本原理,国外医学·物理医学与康复学分册,2005,25(2):53～55
[55] Di Lazzaro,Role of motor evoked potential in diagnoisis of cauda equina and lumbosacral cord lesions,Neurology,2004,63(12):2266~2271
[56]赵强,王明翠,孙霰等,脑诱发电位对酒依赖患者脑干功能及皮质认知功能的评估作用,中国临床康复,2004,8(25):5246～5247
[57]王飞翔张玲莉,勃起功能障碍的诊断方法及其进展,法医学杂志,2005,21(3):216～218
[58]石伟王顺德汪惠书等,神经电生理检测对勃起功能障碍的诊断价值,临床神经电生理学杂志,2004,13(3):168
[59]邵蓓李澄棣王汛,心因性阳痿患者的神经电生理表现,临床神经电生理学杂志,2002,11(3):150～151