Human Brain-Computer Interface
Alexa Riehle, Eilon Vaadia in Motor Cortex in Voluntary Movements, 2004
Several EEG studies indicate that primary sensorimotor areas are activated when subjects imagine the execution of a hand movement. Klass and Bickford43 and Chatrian et al.44 have already observed a blocking or desynchronization of the central mu rhythm with motor imagery. Quantification of the temporal-spatial pattern of ERD clearly showed that one-sided hand motor imagery can result in a lateralized activation of sensorimotor areas, as found in the planning and preparatory phases of a self-paced hand or finger movement.31 Furthermore, measurements of slow cortical potential shifts45 have shown that similar changes over the contralateral hand area can be observed during execution and imagination of movement. The evaluation of movement-related potentials in Parkinson's disease revealed similar potentials during motor imagery and during preparation of a real movement.46 Also, multichannel neuromagnetic measurements demonstrated the effect of motor imagery on brain oscillations generated in primary motor areas.47
ECoG-Based BCIs
Chang S. Nam, Anton Nijholt, Fabien Lotte in Brain–Computer Interfaces Handbook, 2018
In contrast to broadband gamma, low-frequency oscillatory activity provides an index of cortical excitability (Fitzgibbon et al. 2004; Haegens et al. 2011; Howard et al. 2003; Kubanek et al. 2015, 2013; Miltner et al. 1999; Sederberg et al. 2003; Singer and Gray 1995; Szczepanski et al. 2014; Womelsdorf et al. 2006) and plays a central role in the dynamic modulation of cortical function in response to varying task demands (Fries 2005; Jensen and Mazaheri 2010; Schalk 2015). Thus, even though low-frequency activity is likely produced by electrical events in certain (putatively subcortical) populations of neurons, it has proven to be a useful metric of the modulation of the cortex that is different from cortical excitation indexed by broadband gamma. Oscillations at different frequencies subserve different cortical regions. For example, activity in the alpha (8–12 Hz) band is prevalent throughout the sensorimotor system (e.g., Kubanek et al. 2015, 2013) where it is usually referred to as the mu rhythm that is well described in the classical EEG literature (Chatrian 1976) (see Figure 16.5Brunner et al. 2009). Typically, the mu rhythm and the closely associated beta (18–26 Hz) rhythm are relatively focused spectrally and appear as peaks in the power spectrum but are relatively widespread spatially (see Figure 16.5a, bottom). Although their peak amplitude modulates with actual or imagined movements (Crone et al. 1998b; Pfurtscheller and Cooper 1975) (see Figure 16.6Schalk 2006), activity in mu or beta bands appears to reveal only modest information about localized differential cortical processing (Toro et al. 1994). Outside the sensorimotor system, alpha oscillations are also prevalent in the visual system (e.g., Van Dijk et al. 2008) and auditory system (e.g., Potes et al. 2014, 2012). In contrast, oscillations in the theta (4–8 Hz) band are pervasive in prefrontal and hippocampal networks (Anderson et al. 2009; Dürschmid et al. 2014; Fujisawa and Buzsáki 2011). Across all these types of systems and oscillations, oscillatory amplitude is typically large during rest, and reduced while the subject is engaging in corresponding function (e.g., Figure 16.6).
Neurologic Diagnosis
Philip B. Gorelick, Fernando D. Testai, Graeme J. Hankey, Joanna M. Wardlaw in Hankey's Clinical Neurology, 2020
Wave forms/rhythms (Figure 1.27): Alpha waves: 8–13 per second (Hz). “Alpha rhythm” is sinusoidal alpha activity recorded over the occipital regions, present when the eyes are closed, and attenuated by eye opening or mental activity (Figure 1.25, Figure 1.28). Posterior theta background is present in young children, attaining an 8-Hz alpha frequency by 3 years of age. A small proportion of adults have little or no alpha rhythm, and both slow (4–5 Hz) and fast (14–16 Hz) normal variants of alpha rhythm may be seen. Temporal and frontal alpha activity also forms part of the background.Beta waves: 14–22 Hz, low amplitude (10–20 μV) are recorded diffusely, maximal over the frontal and midline regions.Theta waves: 4–7 Hz, are prominent diffusely in children, gradually diminishing during adolescence and adulthood. Theta activity also emerges with normal drowsiness in adults, especially over the temporal or frontotemporal regions.Delta waves: <4 Hz, are a normal finding in early childhood, but not in the normal wakeful adult save for hyperventilation-induced and posterior slowing of youth, 2–4 Hz delta waves that are admixed with and have the same reactivity as alpha rhythm.Mu rhythm: alpha activity often with characteristic arciform appearance like the Greek letter μ or the top of a picket fence, is distributed asymmetrically in the central regions of the head, and attenuates with contralateral limb movement (Figure 1.29).Lambda waves: a waveform resembling the Greek letter lambda λ, that is an evoked response to actively viewing a visual stimulus, generated in the occipital region by saccadic eye movements (Figure 1.30).
The Post-Movement Beta Rebound and Motor-Related Mu Suppression in Children
Published in Journal of Motor Behavior, 2020
Junyi Hao, Wenfeng Feng, Lingli Zhang, Yu Liao
The frequency of mu rhythm increased with age (Berchicci et al., 2011; Hagne, 1972). As shown in Figure 2(C,F), the frequency of the mu rhythm is lower in children than adults. Based on the peak frequency of mu and beta reflected in the cluster-averaged spectrum distribution in adults, children and all participants, the 7–11 Hz and 9–13 Hz ranges were chosen to represent children’s and adults’ mu rhythms respectively. The 15–30 Hz range was chosen to represent all participants’ beta rhythms. These frequency distributions are consistent with those used in numerous previous studies of mu and beta rhythms in children and adults (e.g., Gaetz et al., 2010; Saby, Meltzoff, & Marshall, 2013). The average power in each time window in the mu (7–11 Hz, 9–13 Hz) and beta (15–30 Hz) bands were extracted to compare within and between groups. In order to determine the movement modulation of each power band relative to its own baselines, repeated-measures ANOVA with Greenhouse–Geisser correction across the three time windows (baseline, movement and post-movement) were performed firstly for mu and beta power in each age group. The data with the baseline values subtracted were then compared for mu rhythm and beta rhythm separately to determine the age differences; using 2 groups (children/adults) × 2 time windows (movement/post-movement) mixed repeated-measures ANOVAs with Greenhouse–Geisser correction.
Event-related Desynchronization of Mu Rhythms During Concentric and Eccentric Contractions
Published in Journal of Motor Behavior, 2018
Joo-Hee Park, Heon-Seock Cynn, Kwang Su Cha, Kyung Hwan Kim, Hye-Seon Jeon
MATLAB R2008a (The MathWorks, Natick, MA) was used to analyze the EEG data. Among the 32 electrodes, we selected and analyzed C3 (left hemisphere) and C4 (right hemisphere) because they represent the sensorimotor cortex, and it is known that mu rhythms (8–13 Hz) are measured in those areas of the brain (Marshall, Young, & Meltzoff, 2011; Muthukumaraswamy, Johnson, & McNair, 2004). Attenuation of the mu rhythm has been observed in response to action execution for passive and reflex movements (Chatrian, Petersen, & Lazarete, 1959) and watching others' movements (Gastaut & Bert, 1954). Mu rhythm attenuation represents desynchronization of neural activity, which suggests a significant increase in brain activation (Pfurtscheller, Neuper, Andrew, & Edlinger, 1997). Mu rhythm channels with noise contamination were excluded, and the remaining channels in good condition were averaged and used as a reference. The reference electrode was set by linking the mastoid electrodes, with the ground electrode placed between Fpz and Fz. The EEG signal was filtered using a band-pass filter (0.03–100 Hz) and a notch filter at 60 Hz. Subsequently, the ocular and muscular artifacts were removed using independent component analysis. After removing the artifacts, we performed further single-trial waveform analysis to prevent high frequency signals such as muscular artifacts, which can contaminate EEG data. To process the EEG spectral analysis, we applied continuous wavelet transform with a complex Morlet wavelet to each single-trial EEG (Tallon-Baudry, Bertrand, Delpuech, & Pernier, 1996). The frequency ranged from 1 to 100 Hz.
Evaluating the perspectives of those with severe physical impairments while learning BCI control of a commercial augmentative and alternative communication paradigm
Published in Assistive Technology, 2023
Kevin M. Pitt, Jonathan S. Brumberg
A new high technology-based AAC access for those with severe physical impairments focuses on translating brain signals into communication device control. Noninvasive brain–computer interface access methods for AAC (BCI-AAC) are currently in development and have the potential to be viewed alongside existing AAC access methods as an option for communication device control by those with severe physical impairments (Brumberg et al., 2018). Noninvasive BCI-AAC techniques commonly employ electroencephalography (EEG), which records brain activity observable from the scalp using surface electrodes. Similar to existing AAC access techniques, a range of BCI-AAC methods are in development to provide communication device control, which can broadly be categorized into those targeting brain signals associated with sensory, or motor tasks. For instance, motor-based BCI-AAC techniques translate changes in brain activity associated with imagined or attempted target movements into communication device selections (e.g., Vaughan et al., 2006). In more detail, when the brain is at rest, neurons produce rhythmic and synchronized activity between 8 and 12 Hz known as the alpha rhythm (Pfurtscheller & Da Silva, 1999; Pfurtscheller & Neuper, 2009), which is commonly termed the mu-rhythm when measured over sensorimotor areas of the brain (Kuhlman, 1978). An important property of the mu-rhythm is its change in power when the brain engages in processing information or performing physical or imagined motor tasks. Specifically, as neural synchronization decreases with attempted or imagined movements, then so does the overall power in the mu frequency band (at rest, synchronization and mu power increase; Pfurtscheller & Da Silva, 1999). When decreases in neural synchrony are due to an event (e.g., cued performance of a motor task), it is known as event related desynchronization (Pfurtscheller & Da Silva, 1999), which can be traced back to a specific event for translation into a computer command (e.g., select an item).
Related Knowledge Centers
- Alpha Wave
- Electrocorticography
- Electroencephalography
- Magnetoencephalography
- Mirror Neuron
- Motor Cortex
- Neural Oscillation
- Pyramidal Cell
- Visual Cortex
- Neuron