脑电信号特性分析和特征提取的研究
|
摘要
癫痫是一种常见的中枢神经系统疾病，癫痫的发作在临床上一般呈突发性，容易在病人意识丧失的情况下引发事故，造成残疾或死亡。因此，对即将来临的癫痫发作进行预报具有重要的临床意义。利用脑电信号预报癫痫发作的信号处理方法有很多，但取得较好效果的研究结果基本是基于深部脑电信号的。相对而言，头皮脑电具有不侵入病人身体，无痛苦的优点，在临床上得到普遍应用。但头皮脑电易受干扰，信号强度弱，增加了分析和处理的难度，因而目前还没有理想的基于头皮脑电预报癫痫发作的结果。因此，本文的主要的研究目标是寻找合适的方法解决基于头皮脑电的癫痫发作预报问题。
本文主要以小波变换的多分辨分析理论作为特征提取的工具，与递归神经网络相结合，研究其预报癫痫发作的性能。具体内容如下：
首先，概述了基于脑电信号的癫痫发作预报问题的基本理论、历史和现状，并介绍了已有的不同信号处理方法；
其次，研究了小波变换的基础理论，尤其是对正交小波的分频特性进行了详细分析。综合时频窗口中心、窗宽和频域能量集中程度等性能指标，本文认为daubechies 5小波在频域中相邻尺度间的频率重叠程度最轻，适于本文的具体应用。
第三，采用多分辨率分析将头皮脑电信号分解成不同节律成分，分析癫痫发作间期和发作前脑电信号各节律的绝对小波能谱变化状况。然后从相对小波能谱角度比较癫痫脑电信号与正常脑电信号，分析癫痫脑电信号的异常变化。结果发现，癫痫脑电信号中的慢波增强，α节律受到抑制；
第四，深入分析了递归神经网络中基本Elman网络的结构特点以及解耦扩展卡尔曼滤波算法在该网络训练中的应用。在此基础上，本文对基本Elman网络的权值结构进行了改进，将可变的结构单元前馈权值固定为1。理论分析和仿真实验表明，改进后的Elman网络不仅降低了网络训练过程中的计算复杂度和存储空间需求量，而且提高了网络对高阶动态系统的跟踪性能；
最后，结合小波预处理，本文将改进后的基本Elman网络用于癫痫发作预报。比较以三种不同的信号特征作网络输入所得到的预报结果可知，提取小波变换域α节律能谱包络的方法对本文的数据集具有良好的效果。
Epilepsy is a common intrinsic neurological disorder. Epileptic seizures often occur suddenly, causing disability and mortality sometimes when the patient loses his awareness. Predicting an impending epileptic seizure has obvious clinical importance. There are many EEG-based signal processing methods used to predict seizures, but better research results are mainly based on intracranial EEG recordings. In comparison with intracranial EEG, the examine method of the scalp EEG has been used in clinical diagnosis universally for its nondestructive. However, the scalp EEG signal is well known to be subject to the noise or artifact contaminations, making the analysis more difficulty. Therefore none of the ideal seizure predicting results on scalp EEG is published up to now. Thus the goal of this thesis is to find appropriate methods to predict the incoming epileptic seizures based on scalp EEG effectively.
Multiresolution analysis theory is used to extract appropriate features in this thesis and the recurrent neural network is a classifier. The performances of the synthetical system to predict seizures are as follows:
First, the basic theory, the history and the present of seizure prediction based on EEG signals are reviewed, as well as different signal processing methods.
Second, the basic theory of wavelet transform is studied and the frequency properties of the orthogonal wavelets are analyzed in detail. According to such performances as the center of the window in time-frequency domain, the width of the window and the concentrating degree of the power in frequency domain, this thesis considers that the wavelet daubechies 5 is more suitable for the application, because it has less frequency overlap between adjacent scales in frequency domain.
Third, the scalp EEG signals are decomposed to different rhythm components using multiresolution analysis, and the evolution of the absolute wavelet power spectra of different rhythm in intraictal and preictal stage is explored. Then epileptic EEG and normal EEG are compared by the relative wavelet power spectra to find the abnormities. The result is that the slow wave is enhanced and the a rhythm is weakened hi epileptic EEG signals.
Fourth, the architecture properties of the basic Elman network hi the recurrent neural networks is lucubrated, as well as the network training procedure based on the decoupled extended Kalman filter algorithm. According to the characteristic, this thesis proposes an improved method, forcing the feed-forward weights of the context units hi the Elman network to be unit. Theoretical analyses and simulation results show that such improvement can not only decrease the computational complexity and the storage space requirement of the network training algorithm effectively, but also increase the tracking ability to the high order dynamic system of the Elman network.
Finally, combined with the wavelet preprocessing, the unproved Elman network is
applied to seizure prediction. Three different signal features are used as network inputs. It can be concluded by comparing the results that for these data sets, extracting the envelope of the power spectra of the a rhythm in wavelet transform domain can improve the performance of RNN effectively.
本文主要以小波变换的多分辨分析理论作为特征提取的工具，与递归神经网络相结合，研究其预报癫痫发作的性能。具体内容如下：
首先，概述了基于脑电信号的癫痫发作预报问题的基本理论、历史和现状，并介绍了已有的不同信号处理方法；
其次，研究了小波变换的基础理论，尤其是对正交小波的分频特性进行了详细分析。综合时频窗口中心、窗宽和频域能量集中程度等性能指标，本文认为daubechies 5小波在频域中相邻尺度间的频率重叠程度最轻，适于本文的具体应用。
第三，采用多分辨率分析将头皮脑电信号分解成不同节律成分，分析癫痫发作间期和发作前脑电信号各节律的绝对小波能谱变化状况。然后从相对小波能谱角度比较癫痫脑电信号与正常脑电信号，分析癫痫脑电信号的异常变化。结果发现，癫痫脑电信号中的慢波增强，α节律受到抑制；
第四，深入分析了递归神经网络中基本Elman网络的结构特点以及解耦扩展卡尔曼滤波算法在该网络训练中的应用。在此基础上，本文对基本Elman网络的权值结构进行了改进，将可变的结构单元前馈权值固定为1。理论分析和仿真实验表明，改进后的Elman网络不仅降低了网络训练过程中的计算复杂度和存储空间需求量，而且提高了网络对高阶动态系统的跟踪性能；
最后，结合小波预处理，本文将改进后的基本Elman网络用于癫痫发作预报。比较以三种不同的信号特征作网络输入所得到的预报结果可知，提取小波变换域α节律能谱包络的方法对本文的数据集具有良好的效果。
Epilepsy is a common intrinsic neurological disorder. Epileptic seizures often occur suddenly, causing disability and mortality sometimes when the patient loses his awareness. Predicting an impending epileptic seizure has obvious clinical importance. There are many EEG-based signal processing methods used to predict seizures, but better research results are mainly based on intracranial EEG recordings. In comparison with intracranial EEG, the examine method of the scalp EEG has been used in clinical diagnosis universally for its nondestructive. However, the scalp EEG signal is well known to be subject to the noise or artifact contaminations, making the analysis more difficulty. Therefore none of the ideal seizure predicting results on scalp EEG is published up to now. Thus the goal of this thesis is to find appropriate methods to predict the incoming epileptic seizures based on scalp EEG effectively.
Multiresolution analysis theory is used to extract appropriate features in this thesis and the recurrent neural network is a classifier. The performances of the synthetical system to predict seizures are as follows:
First, the basic theory, the history and the present of seizure prediction based on EEG signals are reviewed, as well as different signal processing methods.
Second, the basic theory of wavelet transform is studied and the frequency properties of the orthogonal wavelets are analyzed in detail. According to such performances as the center of the window in time-frequency domain, the width of the window and the concentrating degree of the power in frequency domain, this thesis considers that the wavelet daubechies 5 is more suitable for the application, because it has less frequency overlap between adjacent scales in frequency domain.
Third, the scalp EEG signals are decomposed to different rhythm components using multiresolution analysis, and the evolution of the absolute wavelet power spectra of different rhythm in intraictal and preictal stage is explored. Then epileptic EEG and normal EEG are compared by the relative wavelet power spectra to find the abnormities. The result is that the slow wave is enhanced and the a rhythm is weakened hi epileptic EEG signals.
Fourth, the architecture properties of the basic Elman network hi the recurrent neural networks is lucubrated, as well as the network training procedure based on the decoupled extended Kalman filter algorithm. According to the characteristic, this thesis proposes an improved method, forcing the feed-forward weights of the context units hi the Elman network to be unit. Theoretical analyses and simulation results show that such improvement can not only decrease the computational complexity and the storage space requirement of the network training algorithm effectively, but also increase the tracking ability to the high order dynamic system of the Elman network.
Finally, combined with the wavelet preprocessing, the unproved Elman network is
applied to seizure prediction. Three different signal features are used as network inputs. It can be concluded by comparing the results that for these data sets, extracting the envelope of the power spectra of the a rhythm in wavelet transform domain can improve the performance of RNN effectively.
引文
[1] Lift B., Echauz J.. Prediction of epileptic seizure. THE LANCET Neurology, 2002, 1(5): 22-30.
[2] Rajna P., Clemens B., Csibri E., et al. Hungarian multicentre epidemiologic study of the warring and initial symptoms (prodrome, aura) of epileptic seizures. Seizure, 1997, 6: 361-368.
[3] Weinand M., Carter L., el-Saadany W., et al. Cerebral blood flow and temporal lobe epileptogenicity. J Neurosurg, 1997, 86: 226-232.
[4] Baumgartner C., Serles W., Leutmezer F., et al. Preictal SPECT in temporal lobe epilepsy: regional cerebral blood flow is increased prior to electroencephalography-seizure onset. J Nucl Med, 1998, 39: 978-982.
[5] Geva A. B., Kerem D. H.. Forecasting generalized epileptic seizures from the EEG signal by wavelet analysis and dynamic unsupervised fuzzy clustering. IEEE Transactions on Biomedical Engineering, 1998, 45(10): 1205-1216.
[6] Petrosian A., Prokhorov D., Homan R., et al. Recurrent neural network based prediction of epileptic seizures in intra-and extracranial EEG. Neurocomputing, 2000, 30: 201-218.
[7] 谭郁玲．临床脑电图与脑电地形图学．北京：人民卫生出版社，1999．
[8] 刘建平，郑崇勋．脑电信号的分析——一种探索大脑功能状态及活动规律的途径．国外医学 生物医学工程分册，1995，18(5)：262-270．
[9] 宦飞，王志中，郑崇勋．基于时频分析检测EEG中癫痫样棘／尖波的方法．生物物理学报，2000， 16(3)：539-546。
[10] 刘慧，丁北生，刘洁，等．局限性癫痫脑电时间序列的三种复杂度计算比较．生物物理学报，1998，14(2)：269-274．
[11] 张彤，杨福生，唐庆玉．脑电中癫痫波的自动检测与分类——一种分层次、多方法整合的途径．中国生物医学工程学报，1998，17(1)：1-11．
[12] Ning TK., Joseph D., Bronzino. Bispectral Analysis of the Rat EEG During Various Vigilance States. IEEE Transactions on Biomedical Engineering, 1989, 36(4): 497-499.
[13] Boualem B., Mostefa M.. A Time-Frequency Approach for Newborn Seizure Detection. IEEE Engineering in Medicine and Biology, 2001, 9/10: 54-64.
[14] Cooper R., Winter A. L., Crow H. J., et al. Comparison of subcortical, cortical, and scalp activity using chronically indwelling electrodes in man. Electroencephalogr. Clin. Neurophysiol. 1965, 18: 217-228.
[15] Viglione SS., Ordon VA., Risch F.. A methodology for detecting ongoing changes in the EEG prior to clinical seizures. 21st Western Institute on Epilepsy. West Huntington Beach, CA. McDonnell Douglas Astronautics Co paper WD1399(A). 1970, Feb 27-28.
[16] Gotrnan J., Ives J., Gloor P., et al. Changes in interictal EEG spiking and seizure occurrence in humans. Epilepsia, 1982, 23: 432-433.
[17] Sherwin I. Interictal-ictal transition in feline penicillin epileptogenic focus. Electroencephalogr Clin Neurophysiol, 1978, 45: 525-534.
[18] Lange H., Lieb J., Engel J., et al. Temporo-spatial patterns of pre-ictal spike activity in human temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol, 1983, 56: 543-555.
[19] Iasemidis L., Sackellares J., Zaveri HP., et al. Phase space topography and the Lyapunov exponent of electrocorticograrns in partial seizures. Brain Topogr, 1990, 2: 187-201.
[20] Iasemidis L., Zaveri H., Sackellares J., et al. Phase space analysis of EEG in temporal lobe epilepsy. IEEE Engineering in Medicine and Biology Society 10th Annual International Conference, New Orleans: 1988, Nov 4-7, 1201-1203.
[21] Iasemidis L., Zaveri H., Sackellares J., et al. Nonlinear dynamics of ECoG data in temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol, 1988, 5: 339.
[22] Elger C., Lehnertz K.. Ictogenesis and chaos. Wolf P, ed. Epileptic seizures and syndromes. Sydney: John Libbey & Company Ltd, 1994: 541-546.
[23] Elger C., Lehnertz K.. Seizure prediction by nonlinear time series analysis of brain electrical activity. Eur J Neurosci, 1998, 10: 786-789.
[24] Mormarm F., Lehnertz K., Andrzejak R., et al. Characterizing preictal states by changes in phase synchronization in intracranial EEG recordings from epilepsy patients. Epilepsia, 2000, 41 (suppl.7): 167.
[25] Martinerie J., Adam C., Le Van Quyen M., et al. Epileptic seizures can be anticipated by non-linear analysis. Nat Med, 1998, 4: 1173-1176.
[26] Litt B., Vachtsevanos G., Echauz J., et al. Method and apparatus for predicting the onset of seizures based on features derived from signals indicative of brain activity. International Patent Application (PCT), PCT WO 00/10455. Aug 24, 1998.
[27] 胡广书，王俊峰．癫痫发作前脑电行为的高阶统计量分析．中国生物医学工程学报，1998，17(2)： 111-117．
[28] 孟欣，徐京华，顾凡及．癫痫病人脑电信号的奇异谱．生物物理学报，2001，17(1)：86-90．
[29] Wieser H. Preictal EEG findings. Epilepsia, 1989, 30: 669.
[30] Katz A., Marks D., McCarthy G., et al. Does interictal spiking change prior to seizures? Electroencephalogr Clin Neurophysiol, 1991, 79: 153-156.
[31] Akay, M.. Biomedical Signal Processing. Academic Press, 1994.
[32] Isak Gath, Claude Feuerstein, Dinh Tuan Pham, et al. On the Tracking of Rapid Dynamic Changes in Seizure EEC. IEEE Transactions on Biomedical Engineering, 1992, 39(9): 952-958.
[33] Qin D., Bryan Houchins, Parten M. E., et al. A Comparison of Techniques for the Prediction of Epileptic Seizures. Eighth IEEE Symposium on Computer-Based Medical Systems, IEEE, 1995: 151-158.
[34] Rogowski Z., Gath I., Bental E.. On the Prediction of Epileptic Seizures. Biological Cybernetics, 1981, 42: 9-15.
[35] Gersch, W., Tharp, B. R.. Spectral regression-amount of information analysis seizures in humans. Quantitative analytic studies in epilepsy. New York: Raven Press, 1976, 509-532.
[36] Da Silva, L., Dijk, A., Smits, H.. Detection of nonstationarities in EEG's using the autoregressive model-an application to EEG's of epileptics. Computerized EEG analysis. Stuttgart, New York: Gustav Fisher, 1975, 180-199.
[37] Gersch, W.. Spectral analysis of EEG's by autoregressive decomposition of tine series. Math. Biosci, 1970, 7: 205-222.
[38] Schiff N., Ladar D., Victor J.. Common dynamics in temporal lobe seizures and absence seizures. Neuroscience, 1999, 91(2): 417-428.
[39] Osorio I., Frei M-G, Wilkinson S.B. Real-time automated detection and quantitative analysis of seizures and short-term prediction of clinical onset. Epilepsia, 1998, 39: 615-627.
[40] Widman C., Bingmann D., Lehnertz K., et al. Reduced signal complexity of intracellular recordings: a precursor for epileptiform activity? Brain Res., 1999, 836(1-2): 156-163.
[41] Kantz H., Schreiber T.. Nonlinear time series analysis. Cambridge, UK: Cambridge University Press, 1997.
[42] Hilborn R. C.. Chaos and nonlinear dynamics: an introduction for scientists and engineers. New York: Oxford University Press, 1994.
[43] Peitgen H., Jürgens H., Saupe D.. Fractals for the classroom part two: complex systems and Mandelbrot set. New York: Springer-Verlag, 1992.
[44] Lehnertz K., Elger C.. Can epileptic seizures be predicted? Evidence from nonlinear time series analysis of brain electrical activity. Phys Rev Lett, 1998, 80: 5019-5022.
[45] Lehnertz K., Andrzejak R., Arnhold J., et al. Nonlinear EEG analysis in epilepsy: its possible use for interictal focus localization, seizure anticipation, and prevention. J Clin Neurophysiol, 2001, 18: 209-222.
[46] Iasemidis L., Sackellares J., inventors; University of Florida, assignee. Seizure warning and prediction. US patent 6, 304, 775. Sept 22, 1999.
[47] Iasemidis L., Sackellares J.. Chaos theory and epilepsy. Neuroscientist, 1996, 2: 118-126.
[48] Esteller R., Vachtsevanos G., Echauz J., et al. Fractal dimension characterizes seizure onset in epileptic patients. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP'99). Phoenix, AZ: 1999, March 15-19, 2343-2346.
[49] Velis D., Kalitzin S., Blanes W., et al. Saturability index increases reliability of correlation dimension calculations for ictal state detection in intracranial EEG recordings. Epilepsia, 2000, 41 (suppl 7): 205.
[50] Takens F.. Detecting strange attractors in turbulence. Dynamical systems and turbulence. Warwick 1980 lecture notes in mathematics 898. Berlin: Springer-Verlag, 1981: 366-381.
[51] Le van Quyen M., Martinerie J., Navarro V., et al. Anticipation of epileptic seizures from standard EEG recordings. THE LANCET, 2001, 357: 183-188.
[52] Puskorius G, Feldkamp L.. Decoupled extended Kalman filter training of feedforward layered networks.. Proceedings of International Joint Conference Neural Networks (IJCNN), Seattle, WA, 1991, 1771-1777.
[53] Puskorius G, Feldkamp L.. Neurocontrol of nonlinear dynamical systems with Kalman filter trained recurrent networks. IEEE Trans. Neural Networks, 1994, 5(2): 279-290.
[54] Kalayci T., Ozdamar O.. Wavelet preprocessing for automated neural network detection of EEG spikes. IEEE Engineering in Medicine and Biology. 1995, March/April, 192-202.
[55] 王宏禹．非平稳随机信号分析与处理．国防工业出版社，1999．
[56] 冉启文．小波变换与分数傅立叶变换理论及应用．哈尔滨工业大学出版社，2001．
[57] Sartoretto F., Ermani M.. Automatic detection of epileptiform activity by single-level wavelet analysis. Clinical Neurophysiology, 1999, 110: 239-249.
[58] 谭善文，秦树人，汤宝平．小波基时频特性及其在分析突变信号中的应用．重庆大学学报(自然科学版)，2001，24(2)：12-17．
[59] Haykin S. Neural Networks: A Comprehensive Foundation. Tsinghua University Press, 2001.
[60] 何玉彬，李新忠．神经网络控制技术及应用．科学出版社，2000．
[61] 陈巍，吴捷．递归神经网络的卡尔曼滤波及分层学习算法．华南理工大学学报(自然科学版)，1998，26(4)：44-48．
[62] 邓志东，孙增圻．多层前馈感知器的高阶序贯非线性Kalman滤波学习算法．控制理论与应用，1994，11(3)：381-384。
[63] Schurmann M., Basar E.. Alpha oscillations shed new light on relation between EEG and single neurons. Neuroscience Research, 1999, 33: 79-80.
[64] Basar E., Yordanova J., Kolev V., et al. Is the alpha rhythm a control parameter for brain responses? Biological Cybernetics, 1997, 76: 471-480.
[65] 许敏光，陆怀秀，菅忠，等．混沌理论在癫痫预报和治疗中的应用．国外医学 生物医学工程分册，2002，25(2)：85-88．
[2] Rajna P., Clemens B., Csibri E., et al. Hungarian multicentre epidemiologic study of the warring and initial symptoms (prodrome, aura) of epileptic seizures. Seizure, 1997, 6: 361-368.
[3] Weinand M., Carter L., el-Saadany W., et al. Cerebral blood flow and temporal lobe epileptogenicity. J Neurosurg, 1997, 86: 226-232.
[4] Baumgartner C., Serles W., Leutmezer F., et al. Preictal SPECT in temporal lobe epilepsy: regional cerebral blood flow is increased prior to electroencephalography-seizure onset. J Nucl Med, 1998, 39: 978-982.
[5] Geva A. B., Kerem D. H.. Forecasting generalized epileptic seizures from the EEG signal by wavelet analysis and dynamic unsupervised fuzzy clustering. IEEE Transactions on Biomedical Engineering, 1998, 45(10): 1205-1216.
[6] Petrosian A., Prokhorov D., Homan R., et al. Recurrent neural network based prediction of epileptic seizures in intra-and extracranial EEG. Neurocomputing, 2000, 30: 201-218.
[7] 谭郁玲．临床脑电图与脑电地形图学．北京：人民卫生出版社，1999．
[8] 刘建平，郑崇勋．脑电信号的分析——一种探索大脑功能状态及活动规律的途径．国外医学 生物医学工程分册，1995，18(5)：262-270．
[9] 宦飞，王志中，郑崇勋．基于时频分析检测EEG中癫痫样棘／尖波的方法．生物物理学报，2000， 16(3)：539-546。
[10] 刘慧，丁北生，刘洁，等．局限性癫痫脑电时间序列的三种复杂度计算比较．生物物理学报，1998，14(2)：269-274．
[11] 张彤，杨福生，唐庆玉．脑电中癫痫波的自动检测与分类——一种分层次、多方法整合的途径．中国生物医学工程学报，1998，17(1)：1-11．
[12] Ning TK., Joseph D., Bronzino. Bispectral Analysis of the Rat EEG During Various Vigilance States. IEEE Transactions on Biomedical Engineering, 1989, 36(4): 497-499.
[13] Boualem B., Mostefa M.. A Time-Frequency Approach for Newborn Seizure Detection. IEEE Engineering in Medicine and Biology, 2001, 9/10: 54-64.
[14] Cooper R., Winter A. L., Crow H. J., et al. Comparison of subcortical, cortical, and scalp activity using chronically indwelling electrodes in man. Electroencephalogr. Clin. Neurophysiol. 1965, 18: 217-228.
[15] Viglione SS., Ordon VA., Risch F.. A methodology for detecting ongoing changes in the EEG prior to clinical seizures. 21st Western Institute on Epilepsy. West Huntington Beach, CA. McDonnell Douglas Astronautics Co paper WD1399(A). 1970, Feb 27-28.
[16] Gotrnan J., Ives J., Gloor P., et al. Changes in interictal EEG spiking and seizure occurrence in humans. Epilepsia, 1982, 23: 432-433.
[17] Sherwin I. Interictal-ictal transition in feline penicillin epileptogenic focus. Electroencephalogr Clin Neurophysiol, 1978, 45: 525-534.
[18] Lange H., Lieb J., Engel J., et al. Temporo-spatial patterns of pre-ictal spike activity in human temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol, 1983, 56: 543-555.
[19] Iasemidis L., Sackellares J., Zaveri HP., et al. Phase space topography and the Lyapunov exponent of electrocorticograrns in partial seizures. Brain Topogr, 1990, 2: 187-201.
[20] Iasemidis L., Zaveri H., Sackellares J., et al. Phase space analysis of EEG in temporal lobe epilepsy. IEEE Engineering in Medicine and Biology Society 10th Annual International Conference, New Orleans: 1988, Nov 4-7, 1201-1203.
[21] Iasemidis L., Zaveri H., Sackellares J., et al. Nonlinear dynamics of ECoG data in temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol, 1988, 5: 339.
[22] Elger C., Lehnertz K.. Ictogenesis and chaos. Wolf P, ed. Epileptic seizures and syndromes. Sydney: John Libbey & Company Ltd, 1994: 541-546.
[23] Elger C., Lehnertz K.. Seizure prediction by nonlinear time series analysis of brain electrical activity. Eur J Neurosci, 1998, 10: 786-789.
[24] Mormarm F., Lehnertz K., Andrzejak R., et al. Characterizing preictal states by changes in phase synchronization in intracranial EEG recordings from epilepsy patients. Epilepsia, 2000, 41 (suppl.7): 167.
[25] Martinerie J., Adam C., Le Van Quyen M., et al. Epileptic seizures can be anticipated by non-linear analysis. Nat Med, 1998, 4: 1173-1176.
[26] Litt B., Vachtsevanos G., Echauz J., et al. Method and apparatus for predicting the onset of seizures based on features derived from signals indicative of brain activity. International Patent Application (PCT), PCT WO 00/10455. Aug 24, 1998.
[27] 胡广书，王俊峰．癫痫发作前脑电行为的高阶统计量分析．中国生物医学工程学报，1998，17(2)： 111-117．
[28] 孟欣，徐京华，顾凡及．癫痫病人脑电信号的奇异谱．生物物理学报，2001，17(1)：86-90．
[29] Wieser H. Preictal EEG findings. Epilepsia, 1989, 30: 669.
[30] Katz A., Marks D., McCarthy G., et al. Does interictal spiking change prior to seizures? Electroencephalogr Clin Neurophysiol, 1991, 79: 153-156.
[31] Akay, M.. Biomedical Signal Processing. Academic Press, 1994.
[32] Isak Gath, Claude Feuerstein, Dinh Tuan Pham, et al. On the Tracking of Rapid Dynamic Changes in Seizure EEC. IEEE Transactions on Biomedical Engineering, 1992, 39(9): 952-958.
[33] Qin D., Bryan Houchins, Parten M. E., et al. A Comparison of Techniques for the Prediction of Epileptic Seizures. Eighth IEEE Symposium on Computer-Based Medical Systems, IEEE, 1995: 151-158.
[34] Rogowski Z., Gath I., Bental E.. On the Prediction of Epileptic Seizures. Biological Cybernetics, 1981, 42: 9-15.
[35] Gersch, W., Tharp, B. R.. Spectral regression-amount of information analysis seizures in humans. Quantitative analytic studies in epilepsy. New York: Raven Press, 1976, 509-532.
[36] Da Silva, L., Dijk, A., Smits, H.. Detection of nonstationarities in EEG's using the autoregressive model-an application to EEG's of epileptics. Computerized EEG analysis. Stuttgart, New York: Gustav Fisher, 1975, 180-199.
[37] Gersch, W.. Spectral analysis of EEG's by autoregressive decomposition of tine series. Math. Biosci, 1970, 7: 205-222.
[38] Schiff N., Ladar D., Victor J.. Common dynamics in temporal lobe seizures and absence seizures. Neuroscience, 1999, 91(2): 417-428.
[39] Osorio I., Frei M-G, Wilkinson S.B. Real-time automated detection and quantitative analysis of seizures and short-term prediction of clinical onset. Epilepsia, 1998, 39: 615-627.
[40] Widman C., Bingmann D., Lehnertz K., et al. Reduced signal complexity of intracellular recordings: a precursor for epileptiform activity? Brain Res., 1999, 836(1-2): 156-163.
[41] Kantz H., Schreiber T.. Nonlinear time series analysis. Cambridge, UK: Cambridge University Press, 1997.
[42] Hilborn R. C.. Chaos and nonlinear dynamics: an introduction for scientists and engineers. New York: Oxford University Press, 1994.
[43] Peitgen H., Jürgens H., Saupe D.. Fractals for the classroom part two: complex systems and Mandelbrot set. New York: Springer-Verlag, 1992.
[44] Lehnertz K., Elger C.. Can epileptic seizures be predicted? Evidence from nonlinear time series analysis of brain electrical activity. Phys Rev Lett, 1998, 80: 5019-5022.
[45] Lehnertz K., Andrzejak R., Arnhold J., et al. Nonlinear EEG analysis in epilepsy: its possible use for interictal focus localization, seizure anticipation, and prevention. J Clin Neurophysiol, 2001, 18: 209-222.
[46] Iasemidis L., Sackellares J., inventors; University of Florida, assignee. Seizure warning and prediction. US patent 6, 304, 775. Sept 22, 1999.
[47] Iasemidis L., Sackellares J.. Chaos theory and epilepsy. Neuroscientist, 1996, 2: 118-126.
[48] Esteller R., Vachtsevanos G., Echauz J., et al. Fractal dimension characterizes seizure onset in epileptic patients. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP'99). Phoenix, AZ: 1999, March 15-19, 2343-2346.
[49] Velis D., Kalitzin S., Blanes W., et al. Saturability index increases reliability of correlation dimension calculations for ictal state detection in intracranial EEG recordings. Epilepsia, 2000, 41 (suppl 7): 205.
[50] Takens F.. Detecting strange attractors in turbulence. Dynamical systems and turbulence. Warwick 1980 lecture notes in mathematics 898. Berlin: Springer-Verlag, 1981: 366-381.
[51] Le van Quyen M., Martinerie J., Navarro V., et al. Anticipation of epileptic seizures from standard EEG recordings. THE LANCET, 2001, 357: 183-188.
[52] Puskorius G, Feldkamp L.. Decoupled extended Kalman filter training of feedforward layered networks.. Proceedings of International Joint Conference Neural Networks (IJCNN), Seattle, WA, 1991, 1771-1777.
[53] Puskorius G, Feldkamp L.. Neurocontrol of nonlinear dynamical systems with Kalman filter trained recurrent networks. IEEE Trans. Neural Networks, 1994, 5(2): 279-290.
[54] Kalayci T., Ozdamar O.. Wavelet preprocessing for automated neural network detection of EEG spikes. IEEE Engineering in Medicine and Biology. 1995, March/April, 192-202.
[55] 王宏禹．非平稳随机信号分析与处理．国防工业出版社，1999．
[56] 冉启文．小波变换与分数傅立叶变换理论及应用．哈尔滨工业大学出版社，2001．
[57] Sartoretto F., Ermani M.. Automatic detection of epileptiform activity by single-level wavelet analysis. Clinical Neurophysiology, 1999, 110: 239-249.
[58] 谭善文，秦树人，汤宝平．小波基时频特性及其在分析突变信号中的应用．重庆大学学报(自然科学版)，2001，24(2)：12-17．
[59] Haykin S. Neural Networks: A Comprehensive Foundation. Tsinghua University Press, 2001.
[60] 何玉彬，李新忠．神经网络控制技术及应用．科学出版社，2000．
[61] 陈巍，吴捷．递归神经网络的卡尔曼滤波及分层学习算法．华南理工大学学报(自然科学版)，1998，26(4)：44-48．
[62] 邓志东，孙增圻．多层前馈感知器的高阶序贯非线性Kalman滤波学习算法．控制理论与应用，1994，11(3)：381-384。
[63] Schurmann M., Basar E.. Alpha oscillations shed new light on relation between EEG and single neurons. Neuroscience Research, 1999, 33: 79-80.
[64] Basar E., Yordanova J., Kolev V., et al. Is the alpha rhythm a control parameter for brain responses? Biological Cybernetics, 1997, 76: 471-480.
[65] 许敏光，陆怀秀，菅忠，等．混沌理论在癫痫预报和治疗中的应用．国外医学 生物医学工程分册，2002，25(2)：85-88．