结合子空间方法和提升小波变换提取事件相关脑电位

我们提出了一种提取事件相关脑电位的复合方法。它用奇异值分解方法将含噪信号分解为噪声子空间和信号子空间,将含噪信号正交投影到信号子空间进行初步除噪,随后将得到的信号进行提升小波变换,对变换结果进行一维小波重构进一步去除噪声,最后提取出ERP成分。介绍了基于奇异值分解的子空间方法和对信号进行提升小波去噪的实现方法。仿真结果表明,结合两种方法提取事件相关脑电信号时,比单独采用其中一种方法的效果要好,并可减少提取事件相关电位所需的实验次数。对实验数据的处理结果表明,该方法的实际处理效果良好。     

基于小波变换和盲源分离算法提取事件相关电位

阐述了用小波分解和盲源分离(blind source separation,BSS)算法结合来去除噪声和干扰提取事件相关电位(event-related potential,ERP).采用小波变换分解ERP,抽取出不同频带的细节信息;由小波系数判断选择多个尺度的子带信号,将它们分别与原始ERP组合进行盲分离,方法是极大化信号时间上的可预测性;将分离的结果进一步叠加平均.两类ERP仿真实验结果表明,本文算法提取出的ERP主要成分波明显,易于辨识,信噪比比较单独运用盲分离算法提取出的结果要好.在应用实例中,有效地增强了ERP的μ波.该算法优点在于减少了刺激次数和波形失真,参数变化范围小,在临床上有很好的应用前景.     

基于小波和卡尔曼平滑的事件相关电位单次提取

本研究提出一种事件相关电位单次提取方法,可有效减少实验次数,并可探索实验之间ERP的变异性。此方法基于小波和卡尔曼平滑,首先利用小波变换考察ERP平均信号的时频特性,根据ERP不同分量出现的时间位置,在不同尺度上选取特定的单次实验ERP小波系数构成观测向量,其为真实ERP小波系数状态向量与噪声之和,然后对观测向量进行卡尔曼平滑,最后对卡尔曼平滑后的小波系数进行小波重构,得到单次提取的ERP信号。仿真实验表明,基于小波和卡尔曼平滑的方法不仅信噪比提高约16~18 dB,优于30次叠加平均、简单小波方法和基于高斯基函数的卡尔曼滤波方法,还可以跟踪ERP的幅度趋势变异性。与基于高斯基函数的卡尔曼滤波方法相比,所提方法降低了计算量。真实脑电ERP提取实验表明本方法较好地从单次记录中提取出了事件相关电位,并可解释ERP因适应和应激引起的趋势变异性。     

基于扩展Infomax ICA的ERP少次提取方法研究

事件相关电位(event related potential,ERP)提取是脑电研究的重点之一,目前临床上主要通过相干平均的方法来获取.由于脑电的非平稳性,使其需要大量重复刺激才能获得,对于受试者极不方便,也不利于ERP的实时检测.本文以反映大脑稀少认知事件的相关电位P300为例,采用扩展Infomax(extended informationmaximization)独立分量分析(independent component analysis,ICA)算法,先滤除眼动、工频干扰,再重构脑电数据,最后经少次叠加即可得到与通常需多次相干平均结果相近的比较满意的P300波形.说明ICA算法在ERP的峰值和潜伏期模式识别上具有较为明显的效果,具有潜在的临床工程应用价值.     

基于快速平稳小波变换的特征提取方法分析fMRI数据

多尺度特征提取方法( MFE)利用离散小波变换,通过只重构有用信号所在小波尺度的方式,有效去除了fMRI数据里的干扰成分.但部分尺度重构会导致混频发生,而离散小波变换的结果对小波选取比较敏感.为此,本研究把有用信号最低频率所对应的小波尺度选为舍弃尺度,利用平稳小波变换矩阵算法代替常规的逐体素算法,用sym2、syml0和sym20等3个小波分析了视觉组块型和事件相关型fMRI试验数据.结果显示,所选舍弃尺度的方法避免了混频现象,3个小波给出了相同的且比SPM8更好的分析结果.在时间消耗上,3个小波基于矩阵算法分析一个被试的组块型数据只耗时61 s,sym2基于逐体素算法耗时42 min,SPM8耗时63 s.因此,所提出的方法是一种实用性更强的MFE方法.     

不同性别首发精神分裂症患者听觉事件相关电位的研究

目的 研究不同性别首发精神分裂症患者听觉事件相关电位(ERP)是否存在差异,并分别探讨两组事件相关电位与临床阴性、阳性症状的关系.方法 对60例(男32例,女28例)首发神分裂症患者及30例正常人进行事件相关电位检测,并运用阳性和阴性症状量表进行精神症状评定.结果 ①两患者组的N2、P3潜伏期明显长于对照组,P3波幅明...     

基于汉英语码转换的ERP信号特征提取研究

本研究的目的是应用事件相关电位(ERP)技术,探索汉英语码转换的机制,以提高汉英双语表达和双语教学的水平。首先,采用一种基于时间间隔因素的语码转换范式和脑电信号采集技术,提取了汉英语码转换的ERP信号;随后,对预处理后的ERP信号提取了5种特征:潜伏期、波峰值、功率谱、小波子带能量和小波熵;其中,前两种特征是大多数语码转换的研究中主要提取的特征,后3种是结合随机信号的特征提取算法提取的新特征;最后,采用特征分布图和三因素重复测量方差分析表对5种特征分析讨论。3种新特征的结果表明,长间隔的N2效应大于短间隔的,且时间间隔因素的零假设成立的概率均小于0.05,主效应达到显著水平;但潜伏期、波峰值的分析中未出现前面现象,说明3种新特征比潜伏期、波峰值更有效。通过分析3种新特征,可得出结论:长间隔的N2效应比短间隔的显著,验证了盛行的抑制假说。     

ERP事件相关成分分解算法研究

当单次实验记录中包含两个或多个事件时,以不同事件时间为参考得到的事件相关电位(ERP)将导致不同事件相关的ERP成分之间形成交叉干扰,从而影响对ERP成分的正确理解。为了恢复真实的事件相关成分,Zhang(1998)提出了一种两个事件成分分解的算法。本研究中首先分析该算法的特点,发现其对边界条件的选取非常敏感,采用周期性边界条件的计算结果明显好于非周期边界条件的计算结果,同时发现该算法对事件时间分布形状并不敏感,仿真结果显示采用高斯分布和均匀分布的事件时间能得到相类似的结果,但算法对事件时间分布的方差也较敏感,方差越大分解的效果越好。其次,扩展两个事件成分分解的算法到3个和多个事件成分的分解。最后应用多成分分解算法到真实实验数据(3事件——提示,刺激,反应),结果显示,事件"提示"引起的ERP波形在提示后大约500ms后回到基线水平,符合心理学过程,并且有助于心理学实验设计。     

视、听感觉刺激双模增进作用研究

目的 研究了视(V)、听(A)单模靶刺激与视听双模联合靶刺激(VA)条件下大脑认知作用的差异,进而讨论了双模增进作用的神经机制.方法 采用64道脑电(EEG)信号采集系统,记录了14个被试者在视觉、听觉和视听联合靶刺激条件下的脑诱发电位和行为学数据.检测了单模与双模间行为学[反应时间(RT)与错误率(ER)]和事件相关电位(ERPs)(P2峰值和峰值潜伏期,P3峰值和峰值潜伏期)结果的差异;分析了行为学与ERPs数据间的相关性.结果 结果显示RT、ER和P3峰值潜伏期在单模与双模间的差异具有统计学意义;P3峰值潜伏期与行为学数据特别是反应时间有显著性相关.结论 通过对比视听双模靶刺激和视、听单模靶刺激的行为学和ERPs结果,提示相对视、听单模靶刺激,视听双模靶刺激不仅在早期ERP成份(P1和N1)而且在晚期ERP成分(P2和P3),都表现出优越的感觉信号神经处理机制.     

急性亚硝酸盐中毒患者ERP的研究

本文19例急性亚硝酸盐中毒患者72小时内进行视觉、听觉事件相关电位(ERP)测定结果表示,ERP中P_3波的潜伏期延长,而此时患者血糖值均降低。随着临床症状的好转,血糖值与P_3波潜伏期均恢复正常且两者有显著的相关性。本文对其机理进行初步探讨。     

Spatiotemporal analysis of experimental differences in event-related potential data with partial least squares

One challenge in the analysis of event-related potentials (ERPs) is to identify task-related differences in scalp topography. The multivariate Partial Least Squares (PLS) analysis was used to identify the spatiotemporal distribution of ERP differences related to experimental manipulations. Two simulations included latency shifts and amplitude changes at peaks with temporal overlap. PLS identified effects only at modeled timepoints and electrodes. In contrast, principal components analysis identified differences at most timepoints. We also demonstrated that PLS identified combinations of waveform differences, not isolated sources. ERP components in an auditory oddball task were also assessed with PLS. The primary distinction was between ERPs on hit and correct rejection trials, expressed at multiple timepoints and electrodes. PLS provides a mechanism to describe experimental differences in ERP waveforms, simultaneously across the head.     

Flanker task with equiprobable congruent and incongruent conditions does not elicit the conflict N2

In many published studies, various modifications of the flanker task have been used. Regardless of the flanker task version, the conflict N2 component has been consistently reported and interpreted as evidence for the resolution of conflict introduced by incongruent flankers. However, ERP studies that used the most basic flanker task (i.e., a version with equiprobable congruent and incongruent conditions in which only congruency between the target and flankers is manipulated) have not provided compelling evidence for the conflict N2 component. We report the results of a large‐sample ERP study employing a basic flanker task that allowed us to revisit the mechanism underlying the resolution of conflict introduced by incongruent flankers. In the behavioral data, we observed the classic effect of congruency. In the ERP data, we found three conflict‐sensitive components: (a) an early frontal component, presumably corresponding to P2, (b) P300 for congruent trials, followed by (c) P300 for incongruent trials. We did not find evidence for the conflict N2 component. Based on a review of literature, we propose that the conflict N2 component observed in a basic flanker task might be a frontal aspect of the P300 component. Given previous attempts to attribute the functional role of the ERP components, the absence of the conflict N2 in the basic flanker task suggests that response inhibition may not be crucial for the resolution of conflict induced by incongruent flankers. Instead, the P2 component appears to indicate that selective attention might play an important role in resolving the flanker conflict.     

The independence of the P300 and the CNV reviewed: a reply to Wastell

The relationship of P300 to the CNV was investigated by Donchin, Tueting, Ritter, Kutas and Heffley (1975) who concluded that these two components of the event-related brain potential (ERP) are independent. Wastell (1979) questioned the validity of the data analysis procedures and of the experimental design used by Donchin et al. In this report we examine Wastell's criticisms and find them to be unfounded. In support of this conclusion we note the differences between principal component analysis and factor analysis. We clarify points about Donchin et al.'s experimental design, and we review evidence for the independence of the P300 and the CNV that has accumulated since 1975.     

Short-term mobilization of processing resources is revealed in the event-related potential

This study investigates whether an occasional effortful improvement of performance, as asked for by a precue, is reflected in event-related potential (ERP) changes. To estimate the limits of possible effort-induced behavioral and ERP changes, we manipulated the time between precue and imperative stimulus (IS; precue interval, PCI). The subjects could, in fact, improve their performance in the effort trials, with all but the shortest PCI. The postcue ERP revealed a fronto-central contingent negative variation (CNV), which was preceded by a frontal positive/occipital negative wave (P2/N2). Both the P2/N2 and the CNV were larger for effort than for standard trials for all PCIs. For the shortest PCI (300 ms), the CNV increase was seen after the IS. The CNV increase for PCIs 600 and 300 began at about 400 ms postcue. The results suggest that effortful performance improvement is associated with prior increase of a frontocentral CNV and a preceding P2/N2. The CNV increase is thought to reflect the activity of a frontal executive process by which additional processing resources can be mobilized on a trial-to-trial basis within less than 500 ms.     

On the relationship between occipital cortex activity and inhibition of return

The present study explored the relationship between inhibition of return (IOR) and visual processes by seeking evidence that IOR and changes in event-related potential (ERP) indices of occipital cortex activity covary in response to experimental manipulation. The presence or absence of a central reorienting event was manipulated within the context of a cue–target experiment. When a reorienting event was presented in the interval between cue and target, IOR was accompanied by reductions in the amplitudes of early occipital ERP peaks on validly cued trials relative to invalidly cued trials. When a reorienting event was not presented, neither IOR nor modulations of the occipital ERP peaks was observed. These results provide strong evidence that IOR arises from changes in occipital visual processing. We propose that IOR arises from a slowing of response-selection processes on validly cued trials due differences in the perceptual input to the decision-making process.     

Understanding the subliminal affective priming effect of facial stimuli: an ERP study

The subliminal affective priming was studied by recording event-related potential (ERP) to ambiguous neutral faces preceded by 20 ms positive or negative prime faces. Behavioral results showed that positive and negative prime faces led to valance-consistent biases in affective judgments of ambiguous neutral faces. ERP data showed that subliminal primes differentially influenced N100, N300 and vertex positive potential (VPP), with larger amplitudes in negative priming trials. These results support the conclusion that the early perceptual analysis of targets, as well as evaluation and selection of targets, is affected by subliminal affective information.     

Taking into account latency, amplitude, and morphology: improved estimation of single-trial ERPs by wavelet filtering and multiple linear regression

Across-trial averaging is a widely used approach to enhance the signal-to-noise ratio (SNR) of event-related potentials (ERPs). However, across-trial variability of ERP latency and amplitude may contain physiologically relevant information that is lost by across-trial averaging. Hence, we aimed to develop a novel method that uses 1) wavelet filtering (WF) to enhance the SNR of ERPs and 2) a multiple linear regression with a dispersion term (MLR(d)) that takes into account shape distortions to estimate the single-trial latency and amplitude of ERP peaks. Using simulated ERP data sets containing different levels of noise, we provide evidence that, compared with other approaches, the proposed WF+MLR(d) method yields the most accurate estimate of single-trial ERP features. When applied to a real laser-evoked potential data set, the WF+MLR(d) approach provides reliable estimation of single-trial latency, amplitude, and morphology of ERPs and thereby allows performing meaningful correlations at single-trial level. We obtained three main findings. First, WF significantly enhances the SNR of single-trial ERPs. Second, MLR(d) effectively captures and measures the variability in the morphology of single-trial ERPs, thus providing an accurate and unbiased estimate of their peak latency and amplitude. Third, intensity of pain perception significantly correlates with the single-trial estimates of N2 and P2 amplitude. These results indicate that WF+MLR(d) can be used to explore the dynamics between different ERP features, behavioral variables, and other neuroimaging measures of brain activity, thus providing new insights into the functional significance of the different brain processes underlying the brain responses to sensory stimuli.     

Decomposing delta, theta, and alpha time-frequency ERP activity from a visual oddball task using PCA.

OBJECTIVE: Time-frequency (TF) analysis has become an important tool for assessing electrical and magnetic brain activity from event-related paradigms. In electrical potential data, theta and delta activities have been shown to underlie P300 activity, and alpha has been shown to be inhibited during P300 activity. Measures of delta, theta, and alpha activity are commonly taken from TF surfaces. However, methods for extracting relevant activity do not commonly go beyond taking means of windows on the surface, analogous to measuring activity within a defined P300 window in time-only signal representations. The current objective was to use a data driven method to derive relevant TF components from event-related potential data from a large number of participants in an oddball paradigm. METHODS: A recently developed PCA approach was employed to extract TF components [Bernat, E. M., Williams, W. J., and Gehring, W. J. (2005). Decomposing ERP time-frequency energy using PCA. Clin Neurophysiol, 116(6), 1314-1334] from an ERP dataset of 2068 17 year olds (979 males). TF activity was taken from both individual trials and condition averages. Activity including frequencies ranging from 0 to 14 Hz and time ranging from stimulus onset to 1312.5 ms were decomposed. RESULTS: A coordinated set of time-frequency events was apparent across the decompositions. Similar TF components representing earlier theta followed by delta were extracted from both individual trials and averaged data. Alpha activity, as predicted, was apparent only when time-frequency surfaces were generated from trial level data, and was characterized by a reduction during the P300. CONCLUSIONS: Theta, delta, and alpha activities were extracted with predictable time-courses. Notably, this approach was effective at characterizing data from a single-electrode. Finally, decomposition of TF data generated from individual trials and condition averages produced similar results, but with predictable differences. Specifically, trial level data evidenced more and more varied theta measures, and accounted for less overall variance.     

基于自适应动态特征库的P300诱发脑电单次提取算法研究

目的 诱发电位的单次提取技术一直是脑电信息处理领域的难题之一,为进一步提高单次提取算法的时间准确性和特征精度,针对体感诱发脑电数据信噪比低、试次间参数变化大的特点,研究诱发脑电参数单次提取新算法,保留试次间诱发脑电的动态特性,并提高估计准确率.方法 基于小波滤波和多元线性分析技术,加入自适应动态特征库并由此提出的诱发脑电P300参数单次提取新方法.随机选取4组小波滤波(WF)后诱发脑电数据,分别叠加平均后进行主成分分析(PCA)组成特征库.单次提取时,针对每试次数据从特征库中选择与当次诱发脑电信号相关系数最高的成分作为自变量开展多元线性回归分析,由回归分析结构重构出单次诱发电位信号并自动提取潜伏期和幅值等关键特征.结果 与专家判定的基准数值相比,新算法预测的P300成分潜伏期与幅值参数更准确,两者的平均差值分别为(11.16±8.60) ms和(1.40±1.34)μV;与常用的叠加平均法结果亦更为接近,平均差值分别为(23.26±25.76) ms和(2.52±2.50) μV,新算法相比传统多元线性回归分析算法具有显著优势.结论 将动态更新的诱发脑电数据主成分样本库应用于小波滤波与多元线性回归方法,能有效保留单次诱发脑电数据中的动态特征,从而提升参数估计的准确率.