基于稀疏表示和外输入自回归模型的单次诱发电位提取方法
|
摘要
由于诱发电位信号的信噪比很小,因此诱发电位的单次提取一直是脑电信号处理领域的难题之一。稀疏表示在信号去噪方面是一个强大的工具,并且诱发电位已被证明在合适的稀疏字典上具有很强的稀疏性。本文提出了一种基于稀疏表示和外输入自回归模型的单次诱发电位提取算法。其中,诱发脑电信号提取的过程分为三个阶段。首先,该算法应用参考信号估计自发脑电的自回归模型参数;其次,在自回归移动平均模型中,应用稀疏表示替代移动平均模型对诱发电位进行建模;最后,利用稀疏系数和自回归模型参数重构诱发电位。通过仿真脑电信号和真实诱发电位信号进行实验验证,结果表明该算法具有较好的效果,与MOSCA和ARX方法相比,本文算法能够在低信噪比情况下进行潜伏期和幅度的估计。
The single-trial extracting of evoked potential( EPs) has become a vexed question in the process of electroencephalogram owing to the faint signal-to-noise ratio( SNR) of EP. Sparse representation is a powerful tool in signal denoising,and EPs have been proven to strong sparsity over an appropriate dictionary. In this paper,we present a novel sparse representation and autoregressive model with exogenous input modeling( ARX) approach to solving the EPs extraction problem. The extraction process of evoked potentials are performed in three stages. First,we utilize an reference ongoing spontaneous Electroencephalogram to identify the parameters of the autoregressive model. Second,instead of the moving average model,sparse representation is used to model the EP in the autoregressive-moving average model. Finally,we calculate the sparse coefficients and derive evoked potentials by using the autoregressive model. Utilizing the synthetic EEG data and real visual evoked potentials data to estimate the performance of this method,our algorithms has been proven highly effective in the experiment results. Comparing with MOSCA and ARX,our method can estimate the latency and amplitude accurately,even in low SNR environments.
The single-trial extracting of evoked potential( EPs) has become a vexed question in the process of electroencephalogram owing to the faint signal-to-noise ratio( SNR) of EP. Sparse representation is a powerful tool in signal denoising,and EPs have been proven to strong sparsity over an appropriate dictionary. In this paper,we present a novel sparse representation and autoregressive model with exogenous input modeling( ARX) approach to solving the EPs extraction problem. The extraction process of evoked potentials are performed in three stages. First,we utilize an reference ongoing spontaneous Electroencephalogram to identify the parameters of the autoregressive model. Second,instead of the moving average model,sparse representation is used to model the EP in the autoregressive-moving average model. Finally,we calculate the sparse coefficients and derive evoked potentials by using the autoregressive model. Utilizing the synthetic EEG data and real visual evoked potentials data to estimate the performance of this method,our algorithms has been proven highly effective in the experiment results. Comparing with MOSCA and ARX,our method can estimate the latency and amplitude accurately,even in low SNR environments.
引文
[1]Reynolds C,Osuagwu B A,Vuckovic A.Influence of motor imagination on cortical activation during functional electrical stimulation[J].Clinical Neurophysiology,2014,126(7):1360-1369.
[2]Fan B,Song H,Liu M,et al.The improvement and realization of speech enhancement algorithm based on Wiener filtering[C]∥International Congress on Image and Signal Processing,2015.
[3]Georgiadis S D,Ranta-Aho P O,Tarvainen M P,et al.Single-trial dynamical estimation of event-related potentials:a Kalman filter-based approach[J].IEEE Transactions on Bio-medical Engineering,2005,52(8):1397-1406.
[4]Wang Z,Roe A W.Trial-to-trial noise cancellation of cortical field potentials in awake macaques by autoregression model with exogenous input(ARX)[J].Journal of Neuroscience Methods,2010,194(2):266-273.
[5]Corbier C,Carmona J.Robust final prediction error criterion for control oriented models validation using L2-L1norm[C]∥International Conference on Communications,Computing and Control Applications,2012:1-6.
[6]Hutapea D K,Yusoff M Z,Asirvadam V S.Single Trial Visual Evoked Potential Extraction using Partial Least Squares-based Approach[J].IEEE Journal of Biomedical&Health Informatics,2014,20(1):82-90.
[7]Wiesel A,Bibi O,Globerson A.Time Varying Autoregressive Moving Average Models for Covariance Estimation[J].IEEE Transactions on Signal Processing,2013,61(11):2791-2801.
[8]De Silva A C,Sinclair N C,Liley D.Limitations in the rapid extraction of evoked potentials using parametric modeling[J].IEEE Transactions on Bio-medical Engineering,2012,59(5):1462-1471.
[9]Causevic E,Morley R E,Wickerhauser M V,et al.Fast wavelet estimation of weak biosignals[J].IEEE Transactions on Biomedical Engineering,2005,52(6):1021-1032.
[10]Markazi S A,Qazi S,Stergioulas L S,et al.Wavelet filtering of the P300 component in event-related potentials[C]∥Conference:International Conference of the IEEE Engineering in Medicine&Biology Society IEEE Engineering in Medicine&Biology Society Conference,2006:1719-1722.
[11]Kreutz-Delgado K,Murray J F,Rao B D,et al.Dictionary Learning Algorithms for Sparse Representation[J].Neural Computation,2006,15(2):349-396.
[12]罗晓宇,王晓庆,甘露,等.张量切片稀疏表示在宽带信号波达方向估计中的应用[J].信号处理,2016,32(8):898-903.Luo Xiaoyu,Wang Xiaoqing,Gan Lu.et al.The Application of Slice-Sparse Representation of Tensor in Directionof-Arrival Estimation for Wideband Sources[J].Journal of Signal Processing,2016,32(8):898-903.(in Chinese)
[13]Xu P,Yao D.Development and evaluation of the sparse decomposition method with mixed over-complete dictionary for evoked potential estimation[J].Computers in Biology&Medicine,2007,37(12):1731-1740.
[14]Yu N,Liu H,Wang X,et al.A joint sparse representation-based method for double-trial evoked potentials estimation[J].Computers in Biology&Medicine,2013,43(12):2071-2078.
[15]胡正平,陈俊岭,王宁,等.先验采样约束结合扩展遮挡字典的细化稀疏人脸识别技术研究[J].信号处理,2015,31(9):1075-1081.Hu Zhengping,Chen Junling,Wang Ning,et al.Detailed Sparse Face Recognition Based on Priori Sampling Constraints and Extended Disguise Dictionary[J].Journal of Signal Processing,2015,31(9):1075-1081.(in Chinese)
[16]胡正平,白帆,王蒙,等.原子-分子字典结合的联合扩展加权稀疏表示人脸识别算法[J].信号处理,2016,32(7):801-809.Hu Zhengping,Bai Fan,Wang Meng,et al.Atom-Molecule Dictionary Joint Extended Weighted Sparse Representation Face Recognition Algorithm[J].Journal of Signal Processing,2016,32(7):801-809.(in Chinese)
[17]Lange D H,Pratt H,Inbar G F.Modeling and estimation of single evoked brain potential components[J].IEEE Transactions on Biomedical Engineering,1997,44(9):791-799.
[18]Lange D H,Inbar G F.A robust parametric estimator for single-trial movement related brain potentials[J].IEEE Transactions on Biomedical Engineering,1996,43(4):341-347.
[19]毕峰,邱天爽,余南南.字典训练结合形态分量分析的诱发电位少次提取方法[J].信号处理,2013,29(3):405-409.Bi Feng,Qiu Tianshuang,Yu Nannan.Fewtrial Extraction of Evoked Potentials with Dictionary Training and Morphological Component Analysis[J].Journal of Signal Processing,2013,29(3):405-409.(in Chinese)
[20]Kamel N,Yusoff M Z,Hani A F.Single-trial subspacebased approach for VEP extraction[J].IEEE Transactions on Bio-medical Engineering,2011,58(5):1383-1393.
[21]Yu N,Hu F,Zou D,et al.Single-Trial Sparse Representation-Based Approach for VEP Extraction[J].Biomed Research International,2016,2016:1-9.
[2]Fan B,Song H,Liu M,et al.The improvement and realization of speech enhancement algorithm based on Wiener filtering[C]∥International Congress on Image and Signal Processing,2015.
[3]Georgiadis S D,Ranta-Aho P O,Tarvainen M P,et al.Single-trial dynamical estimation of event-related potentials:a Kalman filter-based approach[J].IEEE Transactions on Bio-medical Engineering,2005,52(8):1397-1406.
[4]Wang Z,Roe A W.Trial-to-trial noise cancellation of cortical field potentials in awake macaques by autoregression model with exogenous input(ARX)[J].Journal of Neuroscience Methods,2010,194(2):266-273.
[5]Corbier C,Carmona J.Robust final prediction error criterion for control oriented models validation using L2-L1norm[C]∥International Conference on Communications,Computing and Control Applications,2012:1-6.
[6]Hutapea D K,Yusoff M Z,Asirvadam V S.Single Trial Visual Evoked Potential Extraction using Partial Least Squares-based Approach[J].IEEE Journal of Biomedical&Health Informatics,2014,20(1):82-90.
[7]Wiesel A,Bibi O,Globerson A.Time Varying Autoregressive Moving Average Models for Covariance Estimation[J].IEEE Transactions on Signal Processing,2013,61(11):2791-2801.
[8]De Silva A C,Sinclair N C,Liley D.Limitations in the rapid extraction of evoked potentials using parametric modeling[J].IEEE Transactions on Bio-medical Engineering,2012,59(5):1462-1471.
[9]Causevic E,Morley R E,Wickerhauser M V,et al.Fast wavelet estimation of weak biosignals[J].IEEE Transactions on Biomedical Engineering,2005,52(6):1021-1032.
[10]Markazi S A,Qazi S,Stergioulas L S,et al.Wavelet filtering of the P300 component in event-related potentials[C]∥Conference:International Conference of the IEEE Engineering in Medicine&Biology Society IEEE Engineering in Medicine&Biology Society Conference,2006:1719-1722.
[11]Kreutz-Delgado K,Murray J F,Rao B D,et al.Dictionary Learning Algorithms for Sparse Representation[J].Neural Computation,2006,15(2):349-396.
[12]罗晓宇,王晓庆,甘露,等.张量切片稀疏表示在宽带信号波达方向估计中的应用[J].信号处理,2016,32(8):898-903.Luo Xiaoyu,Wang Xiaoqing,Gan Lu.et al.The Application of Slice-Sparse Representation of Tensor in Directionof-Arrival Estimation for Wideband Sources[J].Journal of Signal Processing,2016,32(8):898-903.(in Chinese)
[13]Xu P,Yao D.Development and evaluation of the sparse decomposition method with mixed over-complete dictionary for evoked potential estimation[J].Computers in Biology&Medicine,2007,37(12):1731-1740.
[14]Yu N,Liu H,Wang X,et al.A joint sparse representation-based method for double-trial evoked potentials estimation[J].Computers in Biology&Medicine,2013,43(12):2071-2078.
[15]胡正平,陈俊岭,王宁,等.先验采样约束结合扩展遮挡字典的细化稀疏人脸识别技术研究[J].信号处理,2015,31(9):1075-1081.Hu Zhengping,Chen Junling,Wang Ning,et al.Detailed Sparse Face Recognition Based on Priori Sampling Constraints and Extended Disguise Dictionary[J].Journal of Signal Processing,2015,31(9):1075-1081.(in Chinese)
[16]胡正平,白帆,王蒙,等.原子-分子字典结合的联合扩展加权稀疏表示人脸识别算法[J].信号处理,2016,32(7):801-809.Hu Zhengping,Bai Fan,Wang Meng,et al.Atom-Molecule Dictionary Joint Extended Weighted Sparse Representation Face Recognition Algorithm[J].Journal of Signal Processing,2016,32(7):801-809.(in Chinese)
[17]Lange D H,Pratt H,Inbar G F.Modeling and estimation of single evoked brain potential components[J].IEEE Transactions on Biomedical Engineering,1997,44(9):791-799.
[18]Lange D H,Inbar G F.A robust parametric estimator for single-trial movement related brain potentials[J].IEEE Transactions on Biomedical Engineering,1996,43(4):341-347.
[19]毕峰,邱天爽,余南南.字典训练结合形态分量分析的诱发电位少次提取方法[J].信号处理,2013,29(3):405-409.Bi Feng,Qiu Tianshuang,Yu Nannan.Fewtrial Extraction of Evoked Potentials with Dictionary Training and Morphological Component Analysis[J].Journal of Signal Processing,2013,29(3):405-409.(in Chinese)
[20]Kamel N,Yusoff M Z,Hani A F.Single-trial subspacebased approach for VEP extraction[J].IEEE Transactions on Bio-medical Engineering,2011,58(5):1383-1393.
[21]Yu N,Hu F,Zou D,et al.Single-Trial Sparse Representation-Based Approach for VEP Extraction[J].Biomed Research International,2016,2016:1-9.