基于高频双频SSVEP的脑-机接口编码方法研究

目的通过高频刺激提升基于稳态视觉诱发电位(SSVEP)的脑-机接口(BCI)的用户舒适度,同时结合双频编码,克服高频导致的解码准确率下降问题。方法基于25.5~39.6 Hz频率设计了左右视野和棋盘格视觉刺激的2种60指令双频高频编码范式。共采集了13名受试者的数据,针对SSVEP信号进行频域空域特征分析,并根据频域诱发成分优化滤波器组参数。分别采用滤波器组的扩展典型相关分析(eCCA)、集成任务相关成分分析(eTRCA)以及任务判别模式分析(TDCA)等算法进行SSVEP识别以验证范式可行性。结果左右视野和棋盘格范式均成功诱发了稳定的SSVEP,左右视野基频及其谐波信噪比高,互调成分信噪比较弱,而棋盘格2个刺激频率的互调成分f1+f2的信噪比则明显高于30 Hz以上的二次谐波成分,同时还存在f2?f1成分和2f1?f2成分。结合脑地形图可以看出左右视野的f1和f2响应成分分别位于视野的对侧,而棋盘格则均集中于枕区中央。对于脑地形图振幅和信噪比的偏侧,左右视野刺激频率下PO3和PO4信噪比平均值符合对侧响应特征。5fb?1方法为最优滤波器组设置方法,左右视野TDCA的识别正确率最高,而棋盘格eTRCA和TDCA的识别正确率比较差异没有统计学意义(P>0.05),3种算法的信息传输速率均随数据长度的增加先升高后降低。结论设计的双频高频SSVEP-BCI范式能够较好平衡性能和舒适度,为实用性的大指令集BCI设计方法提供依据。     

咬合动作相关肌电对稳态视觉诱发电位的典型频率影响

目的 基于稳态视觉诱发电位(steady-state visual evoked potential,SSVEP)和肌电(electromyography,EMG)的组合是广泛使用的混合BCI模式.而对于那些只能控制面部肌肉的使用者,咬合动作相关面部肌电通常与SSVEP结合使用.本研究探讨了下颌咬合相关肌电对枕部电极采集到的SSVEP的干扰情况,进而寻找即使在咬合动作下也可同时进行SSVEP识别的刺激频率.方法 根据咬合类型,实验分为3个模式组(无咬合、短咬和长咬合).在每组模式中,受试者同时注视3个闪烁在6.2 Hz、9.8 Hz和14.6 Hz视觉刺激目标.收集枕区4个位点的SSVEP后观察了在3种咬合模式下,各个闪烁刺激的SSVEP响应频谱,并利用典型相关分析方法识别了SSVEP信号,最后统计了准确率.结果 当刺激频率低于20 Hz时,即使有以上2种咬合动作,仍然可以避免其对SSVEP的干扰.根据这些信号的频域特征依然可以识别SSVEP.另外,在咬合动作下进行稳态视觉刺激时,SSVEP的识别率仍然很高(无咬合动作:100.0％,短咬:94.7％,长时间咬合:100.0％).结论 通过合理的频率选择和信号处理,即便下颌咬合动作和SSVEP刺激同时发生时,仍可将咬合动作对稳态视觉诱发电位的影响降低,而且达到较高的识别准确率.     

基于二维集合经验模式分解的稳态视觉诱发电位目标检测研究

稳态视觉诱发电位(SSVEP)是由持续的视觉刺激而诱发的节律性脑电信号。SSVEP频率由固定的视觉刺激频率及其谐波频率组成。二维集合经验模式分解(2D-EEMD)是经典的经验模式分解算法的改进算法,将分解拓展到二维方向上。本文首创性地将2D-EEMD应用于SSVEP。分解得到的本征模式函数(IMF)的二维图像可清晰地观测到SSVEP频率。经过噪声和伪迹滤除的SSVEP主要有效IMF成分投影到头图上,可以反映大脑对视觉刺激的时变趋势,以及大脑不同区域的反应程度,结果显示枕叶区对于视觉刺激的反应最为强烈。最后本文用短时傅里叶变换(STFT)对2D-EEMD的重构信号进行SSVEP频率提取,其识别准确率提高了16%。     

基于不同闪烁频率光刺激的脑电压变化研究

目的：脑电信号是大量神经元电活动在大脑头皮表面产生的电位总和,可综合反映大脑内部功能状态和活跃程度。当人体受到外界刺激时,神经系统产生一系列复杂的生物电活动,最终反映为脑电信号。本文使用不同闪烁频率光为刺激源,对光刺激前后脑电信号时域和频域的变化特性进行分析,为利用外刺激调制脑电频率提供新的实验依据。方法：使用同一光源,分别改变其闪烁频率对受试者眼睛进行刺激,利用脑立方移动版采集不同闪烁频率光刺激下的脑电信号,记录并分析脑电信号电位幅值,及其成份δ波（1 Hz~3 Hz）、θ波（4 Hz~7 Hz）、α波（8 Hz~13 Hz）和β波（14Hz~30Hz）能量的变化情况。结果：随着光闪烁频率的增高,脑电压幅度均值和各频段能量先逐渐增大后减小,脑电幅值和各频段能量均在光闪烁频率为5 Hz时达到最大。结论：不同闪烁频率光刺激对脑电压的变化具有显著性影响：在一定阈值之下,脑电压与光闪烁频率变化一致;超过阈值后,趋势相反。该结论有助于探索脑电信号在外刺激作用下的变化规律,为利用外刺激治疗脑部疾病的方式提供了参考价值。     

视觉感知空间分布对SSVEP调制成分的影响

目的 研究视觉感知空间分布对稳态视觉诱发电位（steady state visual evoked potential,SSVEP）调制成分的影响,为后续研究SSVEP调制成分与认知的关系奠定基础.方法 使用两种闪烁频率分别标记两个刺激图像,设计并实现了“完全重合”、“左右半分”和“对角四分”三种不同空间分布的脑电实验.共9名被试参与此次研究.本文计算并比较了3种实验条件下SSVEP基本成分与调制成分的信噪比空间分布,并做配对t检验.结果 调制成分的信噪比在“完全重合”时远大于“左右半分”和“对角四分”,且在大部分电极处都具有显著性差异.后两种分布下调制成分非常微弱,并且“对角四分”略高于“左右半分”.结论 当不同频率的刺激图像的视觉感知空间交叠时,SSVEP调制成分强,反之,则弱,且可能与交界区域有关.因此在有关SSVEP调制成分的研究中,需考虑不同频率标记的刺激图像的空间分布对其的影响.     

基于Hololens的增强现实脑-机接口研究

当前脑-机接口(BCI)发展迅速,但高性能无创BCI往往需要借助显示设备诱发特定脑电信号,其中最常用的为计算机屏幕,因其难以实现可穿戴而限制BCI的便携性。将增强现实技术(AR)与BCI相结合形成AR-BCI可以解决这一问题,提升BCI的实用性。然而已有AR-BCI研究仅有少量报道,且识别正确率与速度均有待提升。通过采用微软的可穿戴增强现实设备Hololens作为显示设备,实现一种基于稳态视觉诱发电位(SSVEP)的AR-BCI,通过Hololens设备产生视觉刺激诱发八种频率的SSVEP信号,分别开展在线与离线实验,并与基于计算机屏幕的刺激进行比较。参与实验的12名被试成功在增强现实环境中诱发出明显的SSVEP信号,利用1和2 s长的EEG信号分别实现平均88.67%和98.6%的在线识别正确率。该研究表明,AR-BCI有望在日常生活中实现可穿戴的便携化高性能控制型BCI系统。     

不同灰度值刺激产生的稳态视觉诱发电位比较

稳态视觉诱发电位(SSVEP)是大脑对外界光刺激的一种物理反应。本实验从光刺激的物理因素出发,研究不同灰度值刺激对SSVEP频率识别准确率的影响。本实验采用拉普拉斯融合进行空间滤波,并利用典型相关分析法进行频率识别。设置刺激界面背景为白色,改变刺激方块的灰度值,从而分析受试者的频率识别准确率。结果表明灰度值增加,频率识别准确率下降,灰度值为0时,频率识别准确率最高。     

基于增强现实的双眼异频编码稳态视觉诱发电位脑-机接口研究

近年来研究者们将基于稳态视觉诱发电位(SSVEP)的脑-机接口(BCI)系统与增强现实(AR)技术相结合，以提升BCI系统的灵活性和便携性，但基于传统SSVEP范式的AR-BCI普遍性能偏低。本研究利用AR设备对双眼分别进行投影的特点设计了一种双眼异频编码SSVEP范式，通过对左、右眼分别进行不同频率的编码提高SSVEP编码的信息量。共采集了14名被试的数据，采用任务相关成分分析(TRCA)算法进行SSVEP识别，对比双眼异频编码和双眼同频编码SSVEP范式在AR环境下的性能，分析两种编码下SSVEP的频域特征、信噪比和功率谱熵。双眼异频编码采用1、2、3 s脑电数据的平均分类正确率分别为90.9%、93.9%和95.0%,双眼同频编码采用1、2、3 s脑电数据的分类正确率分别为81.1%、87.8%和90.2%。当时间长度小于等于1 s时双眼异频编码刺激的分类正确率显著高于双眼同频编码刺激(0.5 s:t(13)=4.562,P<0.01, Cohen′s d=1.219; 1 s:t(13)=2.737,P<0.05, Cohen′s d=0.732)。特征分析发现，双...     

稳态视觉刺激频率与诱发响应相关性的fMRI研究

稳态视觉诱发电位(SSVEP)及脑电逆问题(inverse EEG)的研究显示了大脑皮层对于不同频率的稳态视觉刺激的响应具有幅度及空间位置上的差异.本文利用功能性磁共振成像(fMRI)研究稳态视觉刺激频率与诱发响应之间的相关性.根据SSVEP的实验结果进行真fMRI实验设计,对5名被试者给予不同频率的稳态视觉刺激,同时进行了脑部的fMRI扫描,采用SPM软件进行数据处理和分析.结果显示,不同频率刺激下的响应区域均集中于初级视皮层(V1),并且均具有单侧优势效应,即右侧半球的响应强于左侧半球.该结果显示了初级视皮层单侧优势对于稳态视觉诱发响应的表现形式.本文讨论了单侧优势与刺激性质之间存在相关性的可能,以及对此做进一步研究和应用的前景.     

结合稳态视觉诱发电位的多模态脑机接口研究进展

脑机接口(BCI)能够在大脑与外部环境之间建立一种不依赖于外周神经或肌肉的交流与控制通道,有助于恢复运动障碍者的生活自理能力,是当前神经工程领域最活跃的研究方向之一。其中,稳态视觉诱发电位(SSVEP)脑机接口因信息传输率(ITR)高、所需训练少而备受关注。现有的无创高通讯速率脑机接口系统主要是来自于或基于SSVEP。近年来,为进一步提高脑机接口性能,整合SSVEP与其他类型输入信号的多模态脑机接口逐渐成为脑机接口研究的新趋势。从输入信号类型、实验范式和信号融合等方面,综述结合SSVEP 的多模态脑机接口研究进展,帮助相关研究者理解该领域研究动态,以启发高通讯速率脑机接口系统的设计与实现。同时探讨目前结合SSVEP的多模态脑机接口存在的问题和未来可能的发展趋势,以期推动结合SSVEP的多模态脑机接口技术的发展。     

Evaluating the feasibility of the steady‐state visual evoked potential (SSVEP) to study temporal attention

Improvements in perceptual performance can be obtained when events in the environment are temporally predictable—and temporal predictability improves attention and sensory processing. The amplitude of the steady‐state visual evoked potential (SSVEP) has been shown to correlate with attention paid to a flickering stimulus even if the flickering stimulus is irrelevant for the task. However, to our knowledge, the validity of the SSVEP to study temporal attention has not been established. Therefore, we designed an SSVEP temporal attention task to evaluate whether the SSVEP and its temporal dynamics can be used to study temporal attention. We used a forced‐choice perceptual detection task while presenting task‐irrelevant visual flicker at alpha (10 Hz) and two surrounding frequencies (6 or 15 Hz). Temporal predictability was manipulated by having the interstimulus intervals (ISI) be constant or variable. Behavioral results replicated previous studies confirming the benefits of temporal expectations on performance for trials with constant ISI. EEG analyses revealed robust SSVEP amplitudes for all flicker frequencies, although a main effect of temporal expectations on SSVEP amplitude was not significant. Additional analyses revealed temporal predictability‐related modulations of SSVEP amplitude at 10 Hz and its second harmonic (20 Hz). The effect of temporal predictability was also observed for the 6 Hz flicker, but not for 15 Hz for any ISI condition. These results provide some evidence for the feasibility of the SSVEP technique to study temporal attention for stimuli with flicker frequencies around the alpha band.     

Changes in BOLD transients with visual stimuli across 1-44 Hz

The dependency of positive BOLD (PBOLD) and post-stimulus undershoot (PSU) on the temporal frequency of visual stimulation was investigated using stimulation frequencies between 1 and 44 Hz. The PBOLD peak at 8 Hz in primary visual cortex was in line with previous neuroimaging studies. In addition to the 8 Hz peak, secondary peaks were observed for stimulation frequencies at 16 and 24 Hz. These additional local peaks were contrary to earlier fMRI studies which reported either a decrease or a plateau for frequencies above 8 Hz but in line with electrophysiological results obtained in animal local field potential (LFP) measurements and human steady-state visual evoked potential (SSVEP) recordings. Our results also indicate that the dependency of PSU amplitude on stimulus frequency deviates from that of PBOLD. Although their amplitudes were correlated within the 1-13 Hz range, they changed independently at stimulation frequencies between 13 and 44 Hz. The different dependency profiles of PBOLD and PSU to stimulation frequency points to different underlying neurovascular mechanisms responsible for the generation of these BOLD transients with regard to their relation to inhibitory and excitatory neuronal activity.     

Clinical feasibility of brain‐computer interface based on steady‐state visual evoked potential in patients with locked‐in syndrome: Case studies

Although the feasibility of brain‐computer interface (BCI) systems based on steady‐state visual evoked potential (SSVEP) has been extensively investigated, only a few studies have evaluated its clinical feasibility in patients with locked‐in syndrome (LIS), who are the main targets of BCI technology. The main objective of this case report was to share our experiences of SSVEP‐based BCI experiments involving five patients with LIS, thereby providing researchers with useful information that can potentially help them to design BCI experiments for patients with LIS. In our experiments, a four‐class online SSVEP‐based BCI system was implemented and applied to four of five patients repeatedly on multiple days to investigate its test‐retest reliability. In the last experiments with two of the four patients, the practical usability of our BCI system was tested using a questionnaire survey. All five patients showed clear and distinct SSVEP responses at all four fundamental stimulation frequencies (6, 6.66, 7.5, 10 Hz), and responses at harmonic frequencies were also observed in three patients. Mean classification accuracy was 76.99% (chance level = 25%). The test‐retest reliability experiments demonstrated stable performance of our BCI system over different days even when the initial experimental settings (e.g., electrode configuration, fixation time, visual angle) used in the first experiment were used without significant modifications. Our results suggest that SSVEP‐based BCI paradigms might be successfully used to implement clinically feasible BCI systems for severely paralyzed patients.     

Competitive effects on steady-state visual evoked potentials with frequencies in- and outside the alpha band

Multiple concurrently presented stimuli are thought to compete for neuronal processing resources. Such competitive stimulus interactions can be investigated by “frequency tagging” each stimulus with an individual temporal frequency. In this case, all stimuli will drive distinct steady-state visual evoked potentials (SSVEPs), hence allowing for an assessment of the distribution of processing resources. Here, we investigated whether competitive effects on SSVEP amplitudes are dependent upon the choice of tagging frequency of either the driving stimulus or a close-by competing stimulus. In particular, we were interested whether changes in amplitude are specific to a 10-Hz SSVEP, as it has been suggested that tagging frequencies within the alpha band drive uniquely characterized neural networks. If this was the case, an additional competition might be introduced when two stimuli are tagged with frequencies within the alpha band and thus compete for processing resources in similar networks. Additionally, we tested whether effects on SSVEP amplitude differ when the competing stimulus is tagged with a frequency of 12 Hz that produces a perceptible flicker when compared to an imperceptible 60-Hz flicker. We found a significant decrease in amplitude of 10- and 15-Hz SSVEPs upon presentation of the competing stimulus regardless of its tagging frequency. Our results clearly indicate that an SSVEP with a frequency within the alpha band and a 15-Hz SSVEP show similar sensitivity to effects of competition. Furthermore, the observed effects of competition on SSVEP amplitude occur independently of flicker perceptibility.     

The Influence of Cognitive Tasks on Different Frequencies Steady-state Visual Evoked Potentials

Previous studies suggested that there exists different neural networks for different frequency bands of steady-state visual evoked potential (SSVEP). What is the effect of the same cognitive task on different frequency SSVEPs? In this work, when a subject was conducting a graded memory task, a 8.3 or 20 Hz flicker was used as a background stimulation. The recorded EEGs were analyzed by the method of steady-state probe topography (SSPT), the results showed that SSVEPs under these two flicker conditions were similar to each other in the various stages of memory process, and were similar to the result of a high alpha band SSVEP as reported before. However, the SSVEP amplitude and latency in the lower frequency band is more clear and stable than that in the higher frequency band. These results suggest that the same cognitive task affects the different frequency SSVEP in a similar way, and the low frequency flicker is a better choice than the high frequency one in such as working memory study.     

An SSVEP-Actuated Brain Computer Interface Using Phase-Tagged Flickering Sequences: A Cursor System

This study presents a new steady-state visual evoked potential (SSVEP)-based brain computer interface (BCI). SSVEPs, induced by phase-tagged flashes in eight light emitting diodes (LEDs), were used to control four cursor movements (up, right, down, and left) and four button functions (on, off, right-, and left-clicks) on a screen menu. EEG signals were measured by one EEG electrode placed at Oz position, referring to the international EEG 10-20 system. Since SSVEPs are time-locked and phase-locked to the onsets of SSVEP flashes, EEG signals were bandpass-filtered and segmented into epochs, and then averaged across a number of epochs to sharpen the recorded SSVEPs. Phase lags between the measured SSVEPs and a reference SSVEP were measured, and targets were recognized based on these phase lags. The current design used eight LEDs to flicker at 31.25 Hz with 45° phase margin between any two adjacent SSVEP flickers. The SSVEP responses were filtered within 29.25–33.25 Hz and then averaged over 60 epochs. Owing to the utilization of high-frequency flickers, the induced SSVEPs were away from low-frequency noises, 60 Hz electricity noise, and eye movement artifacts. As a consequence, we achieved a simple architecture that did not require eye movement monitoring or other artifact detection and removal. The high-frequency design also achieved a flicker fusion effect for better visualization. Seven subjects were recruited in this study to sequentially input a command sequence, consisting of a sequence of eight cursor functions, repeated three times. The accuracy and information transfer rate (mean ± SD) over the seven subjects were 93.14 ± 5.73% and 28.29 ± 12.19 bits/min, respectively. The proposed system can provide a reliable channel for severely disabled patients to communicate with external environments.     

Functional connectivity analysis of steady-state visual evoked potentials

In this paper, functional connectivity of steady-state visual evoked potential (SSVEP) was investigated. Directed transfer function (DTF) was applied to cortical signals recorded from electroencephalography (EEG) in order to obtain connectivity patterns. Flow gain was proposed to assess the role of the specific brain region involved in the information transmission process. We found network connections exist in many regions beyond occipital region. Flow gain mapping both in 8–12 Hz and 13–30 Hz showed that parietal region seemed to serve as the sole hub of information transmission. Further studies of flow gain obtained from channel Pz showed two distinct peaks centered at about 12 Hz low frequency and 20 Hz medium frequency respectively. The low frequency region had a larger value of flow gain. The present study introduced functional connectivity into SSVEP. Furthermore, we put forward the concept of flow gain for the first time to explore the exchange and processing of brain information during SSVEP.     

Optimization of SSVEP brain responses with application to eight-command Brain–Computer Interface

This study pursues the optimization of the brain responses to small reversing patterns in a Steady-State Visual Evoked Potentials (SSVEP) paradigm, which could be used to maximize the efficiency of applications such as Brain–Computer Interfaces (BCI). We investigated the SSVEP frequency response for 32 frequencies (5–84 Hz), and the time dynamics of the brain response at 8, 14 and 28 Hz, to aid the definition of the optimal neurophysiological parameters and to outline the onset-delay and other limitations of SSVEP stimuli in applications such as our previously described four-command BCI system. Our results showed that the 5.6–15.3 Hz pattern reversal stimulation evoked the strongest responses, peaking at 12 Hz, and exhibiting weaker local maxima at 28 and 42 Hz. After stimulation onset, the long-term SSVEP response was highly non-stationary and the dynamics, including the first peak, was frequency-dependent. The evaluation of the performance of a frequency-optimized eight-command BCI system with dynamic neurofeedback showed a mean success rate of 98%, and a time delay of 3.4 s. Robust BCI performance was achieved by all subjects even when using numerous small patterns clustered very close to each other and moving rapidly in 2D space. These results emphasize the need for SSVEP applications to optimize not only the analysis algorithms but also the stimuli in order to maximize the brain responses they rely on.     

Hazardous nature of high-temporal-frequency strobe light stimulation: neural mechanisms revealed by magnetoencephalography

Individuals in contemporary society are continually exposed to various visual stimuli. Such stimulation, especially when high in temporal frequency, may sometimes cause unexpected events such as photosensitive seizures. Although many studies have demonstrated that high-temporal-frequency (>3 Hz) visual stimulation can yield hazardous responses in the CNS, the mechanisms by which it does so are still unclear. We therefore investigated the mechanisms of neural perturbation by high-temporal-frequency strobe light stimulation with high-temporal-frequency resolution (4–20 Hz with an interval of 2 Hz) using magnetoencephalography with high temporal and spatial resolution. We show that (1) three temporal dipole phases (phases 1, 2 and 3, by time course) can be identified in the visual evoked magnetic fields (VEF's) across stimulation frequencies based on the goodness-of-fit values for equivalent current dipole estimation and horizontal dipole directions, (2) the dipole moment of VEF's is correlated with autonomic nervous system activity in phases 1 and 2, (3) some temporal stimulation frequencies enhance magnetic responses in phases 1, 2 and 3, and (4) these frequencies are harmonically related, with a greatest common divisor frequency (fundamental frequency) of approximately 6.5 Hz. Our clarification of the temporal frequency characteristics of VEF's will contribute to understanding of the potential hazardous effects of high-temporal-frequency strobe light stimulation in the CNS.