一种基于EWT-ICEEMDAN的单通道脑电信号眼电伪迹去除算法

脑电信号和眼电信号存在频谱混叠，目前的单通道脑电信号中眼电伪迹去除方法容易造成脑电信号失真。提出一种基于经验小波变换(EWT)和改进的自适应噪声完备经验模态分解(ICEEMDAN)的单通道脑电信号眼电伪迹去除算法。首先使用EWT将单通道脑电信号分解为δ频段和高频段信号，再用ICEEMDAN将δ频段信号自适应分解为多维本征模态函数(IMFs),设置样本熵阈值自动去除眼电伪迹信号，最后重构得到滤波后的脑电信号。基于半模拟脑电数据和真实脑电数据开展实验，结果表明所提算法相比于已有算法能够在去除眼电伪迹的同时更好地保留原始脑电信息。     

能量熵与峰值窗口结合去除眼电伪迹研究

为改进传统独立分量分析自动去除眼电伪迹算法中存在识别眼电分量速度慢、需采集同步参考眼电信号、丢失脑电信号问题,提出一种不需要参考眼电信号的眼电伪迹自动识别去除方法。利用FastICA分解出独立分量,计算各独立分量频谱能量熵,以频谱能量熵值作为判据识别出眼电分量;然后使用峰值窗口分离出眼电分量中存在的脑电信号,与其他独立分量进行拼接;利用FastICA逆变换重构出去眼电伪迹的脑电信号。实验结果表明：该方法能准确快速自动地去除眼电伪迹,并较好地保留其他的脑电信号成分;频谱能量熵识别眼电伪迹平均用时为0.01?s,准确率为98％,适用于实时EOG去除。     

脑电信号中眼电伪迹自动识别与去除方法研究

传统盲源分离算法消除眼电伪迹须用到两个眼电信号作为参考,但在采集眼电信号时易给被试带来不适产生噪声,且识别时需要人为辨别,为了解决这些问题,提出一种基于FastICA的眼电伪迹自动去除方法。该方法先计算出FastICA提取出的各独立成分与GFP（Global Field Power）值的相关系数,再比较相关系数,将其绝对值最大所对应的独立成分识别为眼电伪迹独立成分,最后把该独立成分置零重构干净的脑电信号,实现眼电伪迹的自动去除。通过自采的30例脑电数据实验结果表明：该方法能完全自动地去除眼电伪迹成分并有效保留其他脑电成分,且快速准确,适用于实时场合。     

基于CEEMDAN-ICA的单通道脑电信号眼电伪迹滤除方法

传统盲源分离法不能解决欠定问题,且分离信号与源信号对应关系不确定.提出一种基于自适应噪声完备经验模态分解(CEEMDAN)和独立成分分析(ICA)相结合的脑电信号眼电伪迹自动去除方法.该方法首先将含伪迹脑电信号自适应分解成多维本征模态函数(IMF),以满足盲源分离方法对信号正定或超定要求,再对本征模态函数用ICA方法构建多维源信号,最后利用模糊熵阈值判据判别多维源信号中的伪迹信号,完成滤波并重构脑电信号.该方法相比于其他算法,能更好的去除眼电伪迹并保留原始信息,适合单通道脑电信号预处理.     

脑电信号中眼电伪迹的自动去除算法

为实现眼电伪迹的自动去除,提高算法的有效性和稳健性,提出一种眼电伪迹自动去除算法。采用样本熵和一种通用的伪迹判决方法对眼电伪迹进行自动识别,通过脑电信号的重构实现眼电伪迹的去除。实验结果表明,对于不同长度的真实脑电信号,该算法均能准确地去除眼电伪迹,较好地保留其他的脑电信号成分,且可以完全自动地去除眼电伪迹,适用于实时场合。     

基于K近邻互信息估计的EEG伪迹消除方法

针对脑电信号中的眼电和心电串扰伪迹,提出一种基于最小相依成分分析的互信息(MILCA)算法的伪迹消除方法.在提升小波硬阈值法对多路原始脑电信号去噪基础上,运用MILCA算法对各通道信号进行盲源分离,同时采用信号间互相关系数和互信息量作为指标,分析伪迹分离程度.与Extend Infomax、FastICA 2种常见盲源分离算法的对比结果表明,运用MILCA算法对脑电信号中的眼电及心电伪迹的分离结果最理想.     

基于小波变换和FastICA的眼电伪迹去除研究

为了高效去除脑电信号(Electroencephalogram, EEG)中的眼电伪迹,文章提出一种基于小波变换(Wavelet Transform, WT)和快速独立成分分析(Fast Independent Component Analysis, FastICA)相结合的眼电伪迹去除方法。首先,应用小波变换将信号分解成不同频率的小波分量,采用适合的小波基函数和阈值针对高低频噪声做去噪处理;其次,应用FastICA算法分离出各通道的独立成分,获取纯净的脑电信号;最后,对BCI competition IV公共数据集应用融合算法,并输入支持向量机(Support Vector Machine, SVM)进行分类验证。实验结果表明,相较于单一的小波变换和FastICA算法,采用文章提出的融合算法处理后的脑电信号的SVM分类识别率分别提升了18.9%和15.8%,证明该融合算法对去除脑电信号中的眼电伪迹有较好的效果。     

基于CuBICA算法的EEG伪迹去除方法

在高阶累积量和独立分量分析的基础上,提出一种基于CuBICA算法的脑电信号伪迹去除方法。针对脑电信号中常含有的眼电、心电等伪迹问题,利用小波包方法对原始脑电信号去噪,并进行中心化和白化处理,运用CuBICA算法对消噪后的脑电信号进行盲源分 离。分析分离后各信号间相关性,结果表明,CuBICA算法能成功分离脑电、眼电与心电信号,有效去除纯脑电信号中的各种伪迹。     

结合PCA和JADE的EEG眼电伪迹去除研究

眼电伪迹干扰是脑电信号中的常见干扰,严重影响到有用脑电信号的提取和分析。提出一种基于主分量分析(PCA)和特征矩阵联合相似对角化(JADE)算法相结合的眼电伪迹去除方法,并探讨了主分量分析对伪迹去除的影响。实验结果表明了该算法的有效性及稳健性,并且其时间开销小。此外该算法还可以有效去除其他脑电伪迹及干扰成分。     

改进独立分量算法的眼电伪迹去除方法研究

脑电信号采集过程中易受眼电干扰,给脑电信号分析处理带来极大的不便,由此提出了一种改进独立分量分析（IICA）自动去除眼电伪迹的方法。该方法将水平和垂直眼电信号按照一定的比例混叠成一导新的信号,并与脑电信号一起作为输入;采用基于负熵判据的FastICA算法快速获取各导独立分量;记录此时的负熵判据参数[a],并利用相关系数识别混叠眼电信号独立分量,记录对应的相关系数;[a]加上一定的步长,重复上述步骤至[a]达到阈值时停止;重复多次上述循环,获取均值向量,取出均值向量中最大的相关系数与所对应的[a],根据[a]获取新的独立分量,采用相关系数自动识别混叠眼电独立分量,并置零;再进行ICA逆变换返回到原信号各个电极,即可得到同时去除水平与垂直眼电伪迹后的各导脑电信号。实验结果表明,IICA方法能有效降低去伪迹耗时,极大提高信噪比,减少均方根误差。     

A deep wavelet sparse autoencoder method for online and automatic electrooculographical artifact removal

Electrooculographical (EOG) artifacts are problematic to electroencephalographical (EEG) signal analysis and degrade performance of brain–computer interfaces. A novel, robust deep wavelet sparse autoencoder (DWSAE) method is presented and validated for fully automated EOG artifact removal. DWSAE takes advantage of wavelet transform and sparse autoencoder to become a universal EOG artifact corrector. After being trained without supervision, the sparse autoencoder performs EOG correction on time–frequency coefficients collected after brain wave signal wavelet decomposition. Corrected coefficients are then used for wavelet reconstruction of uncontaminated EEG signals. DWSAE is compared with five other methods: second-order blind identification, information maximization, joint approximation diagonalization of eigen-matrices, wavelet neural network (WNN) and wavelet thresholding (WT). Experimental results on a visual attention task dataset, a mental state recognition dataset and a semi-simulated contaminated EEG dataset show that DWSAE is capable of suppressing EOG artifacts effectively, while preserving the nature of background EEG signals. The mean square error of signals before and after correction by DWSAE on a semi-simulated contaminated EEG segment of 30 s is the lowest (65.62) when compared to the results produced by WNN and WT. DWSAE addresses limitations posed by these methods in three ways. First, DWSAE can be performed automatically and online in a single channel of EEG data; this has advantages over independent component analysis-based methods. Second, its results are robust and stable in comparison with those of other wavelet-based methods. Third, as an unsupervised learning scheme, DWSAE does not require the off-line training that is necessary for WNN and other supervised learning machine learning-based methods.

     

基于独立分量分析的脑电信号的眼电伪迹消除

介绍了独立分量分析技术的基本概念和原理,及其具有代表性的基于负熵最大的快算独立分量分析算法和基于核空间的独立分量分析算法,并分别对脑电中的眼电伪迹进行去除。通过仿真实验表明了独立分量分析算法较快速独立分量分析算法能更好去除眼电伪迹,具有较好准确性和鲁棒性。     

An efficient soft-computing technique for extraction of EEG signal from tainted EEG signal

Electroencephalography (EEG) is the recording of electrical activity of neurons within the brain and is used for the evaluation of brain disorders. But, EEG signals are contaminated with various artifacts which make interpretation of EEGs clinically difficult. In this research paper, we use a soft-computing technique called ANFIS (Adaptive Neuro-Fuzzy Inference System) for the removal of EOG artifact, combined EOG and EMG artifact. Improvement in the output signal to noise ratio and minimum mean square error are used as the performance measures. The outputs of the proposed technique are compared with the outputs of techniques such as neural network, based on ADALINE (Adaptive Linear Neuron) and adaptive filtering method, which makes use of RLS (Recursive Least Squares) algorithm through wavelet transform (RLS-Wavelet). The obtained results show that the proposed method could significantly detect and suppress the artifacts.     

ANC Schemes for the Enhancement of EEG Signals in the Presence of EOG Artifacts

One of the most important applications of adaptive systems is in noise cancellation using adaptive filters. In this paper, we propose adaptive noise cancellation schemes for the enhancement of EEG signals in the presence of EOG artifacts. The effect of two reference inputs is studied on simulated as well as recorded EEG signals and it is found that one reference input is enough to get sufficient minimization of EOG artifacts. This has been verified through correlation analysis also. We use signal to noise ratio and linear prediction spectra, along with time plots, for comparing the performance of the proposed schemes for minimizing EOG artifacts from contaminated EEG signals. Results show that the proposed schemes are very effective (especially the one which employs Newton's method) in minimizing the EOG artifacts from contaminated EEG signals.     

Wavelet-based envelope features with automatic EOG artifact removal: Application to single-trial EEG data

In this study, we propose an analysis system for single-trial classification of electroencephalogram (EEG) data. Combined with automatic EOG artifact removal and wavelet-based amplitude modulation (AM) features, the support vector machine (SVM) classifier is applied to the classification of left finger lifting and resting. Automatic EOG artifact removal is proposed to eliminate the EOG artifacts automatically by means of independent component analysis (ICA) and correlation coefficient. The features are then extracted from the discrete wavelet transform (DWT) data by the AM method. Finally, the SVM is used for the discriminant of wavelet-based AM features. Compared with EEG data without EOG artifact removal, band power features and LDA classifier, the proposed system achieves promising results in classification accuracy.     

Reliable EOG signal-based control approach with EEG signal judgment

This article proposes a reliable EOG signal-based control approach with EEG signal judgment. In this method, raw bio-neurological signals (including EOG and EEG) are first extracted and segmented in the pre-processing stage. The processed bio-neurological signals will then be evaluated by calculating the feature parameters of these signals. Since the feature parameters in bio-neurological signals may be contaminated by various kinds of artifacts, some artifacts of bio-neurological signals can be indicated by means of the feature parameters of bio-neurological signals. Therefore, the bio-neurological signals contaminated with artifacts cannot be adopted to generate control signals or to judge the correctness of control signals. In the proposed method, in order to generate a reliable control signal based on the EOG signal, the EEG signal is adopted to assist in making a judgment about the validity of the EOG signal. With the proposed method, an EOG signal-based control software platform has been implemented. By using this platform, simulation work has been carried out to control the behavior of a robot. The simulation results verified the effectiveness of the proposed method.