FIGURE 3 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
| BCI training system. (A) Illustration of the closed-loop system. This BCI system consists of an EEG amplifier collecting real-time EEG, a PC analyzing EEG signal and providing visual and auditory feedback, and a robot hand providing sensory feedback. (B) The location of EEG electrodes. There were 16 active electrodes placed on the motor and sensory area and a reference electrode placed on the right ear.
Source publication
Objective: Brain-computer interface (BCI) training is becoming increasingly popular in neurorehabilitation. However, around one third subjects have difficulties in controlling BCI devices effectively, which limits the application of BCI training. Furthermore, the effectiveness of BCI training is not satisfactory in stroke rehabilitation. Intermitte...
Contexts in source publication
Context 1
... imagery-based BCI training system was developed as shown in Figure 3A. EEG signals were recorded using 16 active electrodes (g.LADYbird, g.Tec Medical Engineering GmbH, Schiedlberg, Austria). ...
Similar publications
At the moment, neurocomputer interfaces (BCI) make it possible to implement on their basis devices for diagnosing a physical condition, implementing control systems for bionic prostheses, information input means such as neuro chat and character set systems based on brain potentials. At the moment, the main technology for obtaining brain activity fo...
Brain-computer interfaces (BCIs) collect, analyze, and convert brain activity into instructions and send it to the detection system. BCI is becoming popular in under-brain activities in certain conditions such as attention-based tasks. Researchers have recently used combined neuroimaging techniques such as EEG+fNIRS and EEG+fMRI to solve many real-...
An intuitive and generalisable approach to spatial-temporal feature extraction for high-density (HD) functional Near-Infrared Spectroscopy (fNIRS) brain-computer interface (BCI) is proposed, demonstrated here using Frequency-Domain (FD) fNIRS for motor-task classification. Enabled by the HD probe design, layered topographical maps of Oxy/deOxy Haem...
![The placement of sources and detectors to record fNIRS signals [7].](publication/376698049/figure/fig1/AS:11431281213789721@1703204527176/The-placement-of-sources-and-detectors-to-record-fNIRS-signals-7_Q320.jpg)
Functional near-infrared spectroscopy (fNIRS) is a low-cost and noninvasive method to measure the hemodynamic responses of cortical brain activities and has received great attention in brain-computer interface (BCI) applications. In this paper, we present a method based on deep learning and the time-frequency map (TFM) of fNIRS signals to classify...
Functional near-infrared spectroscopy (fNIRS) allows for a reliable assessment of oxygenated blood flow in relevant brain regions. Recent advancements in immersive virtual reality (VR)-based technology have generated many new possibilities for its application, such as in stroke rehabilitation. In this study, we asked whether there is a difference i...
Citations
... Medical applications, particularly in neurorehabilitation [5,6], still remain a predominant use of active brain-computer interfaces (BCIs) as they can improve the quality of life for patients suffering from amputations or paralysis due to neuronal damage, such as stroke [7,8]. Typical interventions range from upper limb rehabilitation [9] to gait enhancement [10], communication support [11], and interactive engagement [12] via the modulation of sensorimotor rhythms to aid motor function restoration and drive neuroplasticity [13]. ...
... PSD features were extracted from several conventional frequency bands, including theta (4-8 Hz), alpha1 (8-10 Hz), alpha2 (10-12 Hz), beta1 (12-21 Hz), beta2 (21)(22)(23)(24)(25)(26)(27)(28)(29)(30), theta to beta , and delta to gamma (0-50 Hz), for each EEG electrode. In addition to the PSD features, we also extracted power ratios, notably alpha (8-12 Hz) to beta (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) and theta to beta ratios, for each electrode. Subsequent to the extraction process, all features underwent log-scaling. ...
... The process involved the decomposition of EEG signals into five spectral bands utilizing a filter bank (FB). The bands selected for this study included the conventional delta (1-4 Hz), theta (4-8 Hz), alpha (8-12 Hz), beta (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30), and gamma (30-50 Hz), given their established roles in representing neuronal dynamics [57]. We used a zero-phase FIR filter to execute the signal filtering, which consisted of two successive filtering steps in opposing directions on a mirror-padded version of the input signal. ...
Brain–computer interface (BCI) technology has emerged as an influential communication tool with extensive applications across numerous fields, including entertainment, marketing, mental state monitoring, and particularly medical neurorehabilitation. Despite its immense potential, the reliability of BCI systems is challenged by the intricacies of data collection, environmental factors, and noisy interferences, making the interpretation of high-dimensional electroencephalogram (EEG) data a pressing issue. While the current trends in research have leant towards improving classification using deep learning-based models, our study proposes the use of new features based on EEG amplitude modulation (AM) dynamics. Experiments on an active BCI dataset comprised seven mental tasks to show the importance of the proposed features, as well as their complementarity to conventional power spectral features. Through combining the seven mental tasks, 21 binary classification tests were explored. In 17 of these 21 tests, the addition of the proposed features significantly improved classifier performance relative to using power spectral density (PSD) features only. Specifically, the average kappa score for these classifications increased from 0.57 to 0.62 using the combined feature set. An examination of the top-selected features showed the predominance of the AM-based measures, comprising over 77% of the top-ranked features. We conclude this paper with an in-depth analysis of these top-ranked features and discuss their potential for use in neurophysiology.
... Some authors referred to motor training applications as rehabilitative or restorative BCIs, to contrast these with assistive BCI for those who lack movement capabilities. One very recent study included here [21] did not use either term (neurofeedback or BCI) but instead used the term "brain state-dependent stimulation" to describe their system for retraining motor function that did not include the BCI component in this case to illustrate that it is the precisely timed afferent volley that is the essential [90,91,93]. This is similar to a recent review by Mane et al. [46] on BCI application for stroke rehabilitation, in which 47 of 50 studies used EEG alone with one using EEG plus MEG [94], one used MEG [95] and one used fNIRS [90]. ...
Background
Brain–computer interfaces (BCI), initially designed to bypass the peripheral motor system to externally control movement using brain signals, are additionally being utilized for motor rehabilitation in stroke and other neurological disorders. Also called neurofeedback training, multiple approaches have been developed to link motor-related cortical signals to assistive robotic or electrical stimulation devices during active motor training with variable, but mostly positive, functional outcomes reported. Our specific research question for this scoping review was: for persons with non-progressive neurological injuries who have the potential to improve voluntary motor control, which mobile BCI-based neurofeedback methods demonstrate or are associated with improved motor outcomes for Neurorehabilitation applications?
Methods
We searched PubMed, Web of Science, and Scopus databases with all steps from study selection to data extraction performed independently by at least 2 individuals. Search terms included: brain machine or computer interfaces, neurofeedback and motor; however, only studies requiring a motor attempt, versus motor imagery, were retained. Data extraction included participant characteristics, study design details and motor outcomes.
Results
From 5109 papers, 139 full texts were reviewed with 23 unique studies identified. All utilized EEG and, except for one, were on the stroke population. The most commonly reported functional outcomes were the Fugl-Meyer Assessment (FMA; n = 13) and the Action Research Arm Test (ARAT; n = 6) which were then utilized to assess effectiveness, evaluate design features, and correlate with training doses. Statistically and functionally significant pre-to post training changes were seen in FMA, but not ARAT. Results did not differ between robotic and electrical stimulation feedback paradigms. Notably, FMA outcomes were positively correlated with training dose.
Conclusion
This review on BCI-based neurofeedback training confirms previous findings of effectiveness in improving motor outcomes with some evidence of enhanced neuroplasticity in adults with stroke. Associative learning paradigms have emerged more recently which may be particularly feasible and effective methods for Neurorehabilitation. More clinical trials in pediatric and adult neurorehabilitation to refine methods and doses and to compare to other evidence-based training strategies are warranted.
... Therefore, fNIRS has high application potential in the field of neuromodulation rehabilitation. Ding et al. (2021) found that BCI training improved brain functional connectivity between motor cortex and prefrontal cortex via fNIRS. Kaiser et al. (2014) also determined that MI-BCI training increased the cortical activation of the supplementary motor cortex (SMA) and the primary motor cortex. ...
Background
The motor imagery brain computer interface (MI-BCI) is now available in a commercial product for clinical rehabilitation. However, MI-BCI is still a relatively new technology for commercial rehabilitation application and there is limited prior work on the frequency effect. The MI-BCI has become a commercial product for clinical neurological rehabilitation, such as rehabilitation for upper limb motor dysfunction after stroke. However, the formulation of clinical rehabilitation programs for MI-BCI is lack of scientific and standardized guidance, especially limited prior work on the frequency effect. Therefore, this study aims at clarifying how frequency effects on MI-BCI training for the plasticity of the central nervous system.
Methods
Sixteen young healthy subjects (aged 22.94 ± 3.86 years) were enrolled in this randomized clinical trial study. Subjects were randomly assigned to a high frequency group (HF group) and low frequency group (LF group). The HF group performed MI-BCI training once per day while the LF group performed once every other day. All subjects performed 10 sessions of MI-BCI training. functional near-infrared spectroscopy (fNIRS) measurement, Wolf Motor Function Test (WMFT) and brain computer interface (BCI) performance were assessed at baseline, mid-assessment (after completion of five BCI training sessions), and post-assessment (after completion of 10 BCI training sessions).
Results
The results from the two-way ANOVA of beta values indicated that GROUP, TIME, and GROUP × TIME interaction of the right primary sensorimotor cortex had significant main effects [GROUP: F(1,14) = 7.251, P = 0.010; TIME: F(2,13) = 3.317, P = 0.046; GROUP × TIME: F(2,13) = 5.676, P = 0.007]. The degree of activation was affected by training frequency, evaluation time point and interaction. The activation of left primary sensory motor cortex was also affected by group (frequency) (P = 0.003). Moreover, the TIME variable was only significantly different in the HF group, in which the beta value of the mid-assessment was higher than that of both the baseline assessment (P = 0.027) and post-assessment (P = 0.001), respectively. Nevertheless, there was no significant difference in the results of WMFT between HF group and LF group.
Conclusion
The major results showed that more cortical activation and better BCI performance were found in the HF group relative to the LF group. Moreover, the within-group results also showed more cortical activation after five sessions of BCI training and better BCI performance after 10 sessions in the HF group, but no similar effects were found in the LF group. This pilot study provided an essential reference for the formulation of clinical programs for MI-BCI training in improvement for upper limb dysfunction.
... Indeed, the time required for TMS mapping precludes studies of early corticomotor reorganisation. Early ('short-term') reorganisation describes neuroplastic changes occurring within minutes of the introduction of a novel stimulus (Björkman, Weibull, Rosén, Svensson, & Lundborg, 2009;Ding et al., 2021). Traditional TMS mapping often requires over 1 hour per assessment, such that accurately capturing this early, and potentially transient, corticomotor reorganisation is not possible. ...
... Importantly, the time required for TMS mapping precludes studies of early corticomotor reorganisation. Early reorganisation describes neuroplastic changes occurring immediately or within minutes of the introduction of a novel stimulus (Björkman et al., 2009;Ding et al., 2021). The temporal profile of this reorganisation is difficult to explore with TMS mapping as it is often not feasible to collect multiple maps over the course of a single experimental session (Miranda et al., 1997). ...
The overarching aim of this thesis was to enhance our understanding of early corticomotor reorganisation in response to novel stimuli (motor skill training and acute pain). To achieve this aim, four primary studies (Chapters 2-5) were conducted and published. Study 1 (Chapter 2) explored the within- and between-session reliability of corticomotor outcomes assessed using rapid TMS mapping (map area, volume, centre of gravity, discrete peaks in corticomotor excitability, and mean motor evoked potential). This study also assessed the validity of rapid mapping by testing its equivalence with traditional mapping methods. Study 2 (Chapter 3) used rapid mapping to investigate corticomotor reorganisation during short-term motor skill learning in thirty individuals. This study demonstrated, for the first time, that reorganisation of lower back muscle representations occurs rapidly (within minutes) in certain individuals. Study 3 (Chapter 4) explored the temporal profile and variability of corticomotor reorganisation in response to acute experimental pain. Findings of this study suggest that early corticomotor responses could be used as an index to predict symptom severity. This could have utility in stratifying individuals according to their likelihood of increased or persistent pain and the development of targeted management strategies. Study 4 (Chapter 5) investigated this possibility using repeated intramuscular injection of nerve growth factor, a novel and clinically-relevant model of musculoskeletal pain. The findings of this study suggest that early rTMS over M1 may expedite recovery following acute musculoskeletal pain or injury. Taken together, this thesis makes a substantial and original contribution to our understanding of neuroplasticity. By evaluating rapid TMS mapping, early corticomotor reorganisation can now be assessed validly and reliably, allowing exploration of early drivers of nervous system plasticity. Decreasing map acquisition times may also increase the utility of TMS beyond research settings, potentially allowing corticomotor reorganisation to be assessed in clinical environments. The experimental studies throughout this thesis provide valuable insight into the temporal profile and modifiability of early corticomotor reorganisation. This work highlights the prognostic and therapeutic utility of exploring early corticomotor reorganisation and the need for further research in this area.
... Neural effects of iTBS are typically investigated by motor evoked potentials (MEP), which are muscular responses elicited by single-pulse TMS (Talelli et al., 2007;Di Lazzaro et al., 2008;Ding et al., 2021b). However, this approach is not applicable to stroke survivors in whom MEPs are not elicitable. ...
Objective:
Intermittent theta burst stimulation (iTBS) has been widely used as a neural modulation approach in stroke rehabilitation. Concurrent use of transcranial magnetic stimulation and electroencephalography (TMS-EEG) offers a chance to directly measure cortical reactivity and oscillatory dynamics and allows for investigating neural effects induced by iTBS in all stroke survivors including individuals without recordable MEPs. Here, we used TMS-EEG to investigate aftereffects of iTBS following stroke.
Methods:
We studied 22 stroke survivors (age: 65.2 ± 11.4 years; chronicity: 4.1 ± 3.5 months) with upper limb motor deficits. Upper-extremity component of Fugl-Meyer motor function assessment and action research arm test were used to measure motor function of stroke survivors. Stroke survivors were randomly divided into two groups receiving either Active or Sham iTBS applied over the ipsilesional primary motor cortex. TMS-EEG recordings were performed at baseline and immediately after Active or Sham iTBS. Time and time-frequency domain analyses were performed for quantifying TMS-evoked EEG responses.
Results:
At baseline, natural frequency was slower in the ipsilesional compared with the contralesional hemisphere (P = 0.006). Baseline natural frequency in the ipsilesional hemisphere was positively correlated with upper limb motor function following stroke (P = 0.007). After iTBS, natural frequency in the ipsilesional hemisphere was significantly increased (P < 0.001).
Conclusions:
This is the first study to investigate the acute neural adaptations after iTBS in stroke survivors using TMS-EEG. Our results revealed that natural frequency is altered following stroke which is related to motor impairments. iTBS increases natural frequency in the ipsilesional motor cortex in stroke survivors. Our findings implicate that iTBS holds the potential to normalize natural frequency in stroke survivors, which can be utilized in stroke rehabilitation.
... Another limitation of our study is we stimulated the right M1 instead of the left M1, contrary to previous studies investigating the after-effect of iTBS on M1 (Corp et al. 2020). Nevertheless, several iTBS studies targeted the right M1 to assess the plasticity of M1 (Suppa et al. 2008;Platz et al. 2018;Ding et al. 2021). Furthermore, Suppa and colleagues reported that performing iTBS over M1 of the dominant or nondominant hemisphere did not significantly affect the MEP amplitude change on the contralateral FDI muscle (Suppa et al. 2008). ...
Intermittent theta burst stimulation (iTBS) delivered by transcranial magnetic stimulation (TMS) produces a long-term potentiation (LTP)-like after-effect useful for investigations of cortical function and of potential therapeutic value. However, the iTBS after-effect over the primary motor cortex (M1) as measured by changes in motor evoked potential (MEP) amplitude exhibits a largely unexplained variability across individuals. Here, we present evidence that individual differences in white and gray matter microstructural properties revealed by fractional anisotropy (FA) predict the magnitude of the iTBS-induced after-effect over M1. The MEP amplitude change in the early phase (5–10 min post-iTBS) was associated with FA values in white matter tracts such as right superior longitudinal fasciculus and corpus callosum. In contrast, the MEP amplitude change in the late phase (15–30 min post-iTBS) was associated with FA in gray matter, primarily in right frontal cortex. These results suggest that the microstructural properties of regions connected directly or indirectly to the target region (M1) are crucial determinants of the iTBS after-effect. FA values indicative of these microstructural differences can predict the potential effectiveness of rTMS for both investigational use and clinical application.
... As iTBS employs a shorter stimulation period and a lower stimulation intensity compared with traditional rTMS, iTBS could be a good rTMS option in clinical practice (Talelli et al., 2007). Neural effects of iTBS are typically investigated by motor evoked potentials (MEP), which are muscular responses elicited by single-pulse TMS (Huang et al., 2005;Talelli et al., 2007;Di Lazzaro et al., 2008;Ackerley et al., 2010;Hinder et al., 2014;Ding et al., 2021b). However, this approach is not applicable to stroke survivors in whom MEPs in the paretic limb cannot be elicited. ...
Objective: Intermittent theta burst stimulation (iTBS) is a special form of repetitive transcranial magnetic stimulation (rTMS), which effectively increases cortical excitability and has been widely used as a neural modulation approach in stroke rehabilitation. As effects of iTBS are typically investigated by motor evoked potentials, how iTBS influences functional brain network following stroke remains unclear. Resting-state electroencephalography (EEG) has been suggested to be a sensitive measure for evaluating effects of rTMS on brain functional activity and network. Here, we used resting-state EEG to investigate the effects of iTBS on functional brain network in stroke survivors.
Methods: We studied thirty stroke survivors (age: 63.1 ± 12.1 years; chronicity: 4.0 ± 3.8 months; UE FMA: 26.6 ± 19.4/66) with upper limb motor dysfunction. Stroke survivors were randomly divided into two groups receiving either Active or Sham iTBS over the ipsilesional primary motor cortex. Resting-state EEG was recorded at baseline and immediately after iTBS to assess the effects of iTBS on functional brain network.
Results: Delta and theta bands interhemispheric functional connectivity were significantly increased after Active iTBS (P = 0.038 and 0.011, respectively), but were not significantly changed after Sham iTBS (P = 0.327 and 0.342, respectively). Delta and beta bands global efficiency were also significantly increased after Active iTBS (P = 0.013 and 0.0003, respectively), but not after Sham iTBS (P = 0.586 and 0.954, respectively).
Conclusion: This is the first study that used EEG to investigate the acute neuroplastic changes after iTBS following stroke. Our findings for the first time provide evidence that iTBS modulates brain network functioning in stroke survivors. Acute increase in interhemispheric functional connectivity and global efficiency after iTBS suggest that iTBS has the potential to normalize brain network functioning following stroke, which can be utilized in stroke rehabilitation.
Aims
Understanding the neural mechanisms underlying stroke recovery is critical to determine effective interventions for stroke rehabilitation. This study aims to systematically explore how recovery mechanisms post‐stroke differ between individuals with different levels of functional integrity of the ipsilesional corticomotor pathway and motor function.
Methods
Eighty‐one stroke survivors and 15 age‐matched healthy adults participated in this study. We used transcranial magnetic stimulation (TMS), electroencephalography (EEG), and concurrent TMS‐EEG to investigate longitudinal neurophysiological changes post‐stroke, and their relationship with behavioral changes. Subgroup analysis was performed based on the presence of paretic motor evoked potentials and motor function.
Results
Functional connectivity was increased dramatically in low‐functioning individuals without elicitable motor evoked potentials (MEPs), which showed a positive effect on motor recovery. Functional connectivity was increased gradually in higher‐functioning individuals without elicitable MEP during stroke recovery and influence from the contralesional hemisphere played a key role in motor recovery. In individuals with elicitable MEPs, negative correlations between interhemispheric functional connectivity and motor function suggest that the influence from the contralesional hemisphere may be detrimental to motor recovery.
Conclusion
Our results demonstrate prominent clinical implications for individualized stroke rehabilitation based on both functional integrity of the ipsilesional corticomotor pathway and motor function.
The electroencephalogram (EEG) signal from motor imagery (MI) is used to drive brain-computer interaction (BCI). However, users usually are not adept at performing MI, which leads to low quality EEG signals and decreases the performance of BCI applications. The humanoid robot stimulation approach can guide users in performing MI more proficiently by increasing the cortico-spinal excitability and improving the discrimination of ERD patterns during MI tasks. Compared to the traditional stimulation modes, our proposed humanoid robot stimulation mode can activate higher-quality MI EEG signals. We use CNN and LSTM algorithm for extraction of EEG features and classification. The results showed that the CNN-LSTM can achieve the highest classification accuracy (93.7% ±1.7%) in humanoid robot stimulation mode, and it outperformed all other classifierstimulation mode combinations. This demonstrates the effectiveness and feasibility of using a humanoid robot in realscene MI-BCI application, such as service robots or rehabilitation system for person with motor disabilities.
