Search

The use of Riemannian geometry decoding algorithms in classifying electroencephalography-based motor-imagery brain–computer interfaces (BCIs) trials is relatively new and promises to outperform the current state-of-the-art methods by overcoming the noise and nonstationarity of electroencephalography signals. However, the related literature shows high classification accuracy on only relatively small BCI datasets. The aim of this paper is to provide a study of the performance of a novel implementation of the Riemannian geometry decoding algorithm using large BCI datasets. In this study, we apply several Riemannian geometry decoding algorithms on a large offline dataset using four adaptation strategies: baseline, rebias, supervised, and unsupervised. Each of these adaptation strategies is applied in motor execution and motor imagery for both scenarios 64 electrodes and 29 electrodes. The dataset is composed of four-class bilateral and unilateral motor imagery and motor execution of 109 subjects. We run several classification experiments and the results show that the best classification accuracy is obtained for the scenario where the baseline minimum distance to Riemannian mean has been used. The mean accuracy values up to 81.5% for motor execution, and up to 76.4% for motor imagery. The accurate classification of EEG trials helps to realize successful BCI applications that allow effective control of devices.

In motor imagery (MI) brain–computer interface (BCI) research, some researchers have designed MI paradigms of force under a unilateral upper-limb static state. It is difficult to apply these paradigms to the dynamic force interaction process between the robot and the patient in a brain-controlled rehabilitation robot system, which needs to induce thinking states of the patient’s demand for assistance. Therefore, in our research, according to the movement of wiping the table in human daily life, we designed a three-level-force MI paradigm under a unilateral upper-limb dynamic state. Based on the event-related de-synchronization (ERD) feature analysis of the electroencephalography (EEG) signals generated by the brain’s force change motor imagination, we proposed a multi-scale temporal convolutional network with attention mechanism (MSTCN-AM) algorithm to recognize ERD features of MI-EEG signals. Aiming at the slight feature differences of single-trial MI-EEG signals among different levels of force, the MSTCN module was designed to extract fine-grained features of different dimensions in the time–frequency domain. The spatial convolution module was then used to learn the area differences of space domain features. Finally, the attention mechanism dynamically weighted the time–frequency–space domain features to improve the algorithm’s sensitivity. The results showed that the accuracy of the algorithm was 86.4 ± 14.0% for the three-level-force MI-EEG data collected experimentally. Compared with the baseline algorithms (OVR-CSP+SVM (77.6 ± 14.5%), Deep ConvNet (75.3 ± 12.3%), Shallow ConvNet (77.6 ± 11.8%), EEGNet (82.3 ± 13.8%), and SCNN-BiLSTM (69.1 ± 16.8%)), our algorithm had higher classification accuracy with significant differences and better fitting performance.

In the last decades, the automatic recognition and interpretation of brain waves acquired by electroencephalographic (EEG) technologies have undergone remarkable growth, leading to a consequent rapid development of Brain Computer Interfaces (BCIs). EEG-based BCIs are non-invasive systems that allow communication between a human being and an external device interpreting brain activity directly. Thanks to the advances in neurotechnologies, and especially in the field of wearable devices, BCIs are now also employed outside medical and clinical applications. Within this context, this paper proposes a systematic review of EEG-based BCIs, focusing on one of the most promising paradigms based on Motor Imagery (MI) and limiting the analysis to applications that adopt wearable devices. This review aims to evaluate the maturity levels of these systems, both from the technological and computational points of view. The selection of papers has been performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), leading to 84 publications considered in the last ten years (from 2012 to 2022). Besides technological and computational aspects, this review also aims at systematically list experimental paradigms and available datasets in order to identify benchmarks and guidelines for the development of new applications and computational models.

This review article discusses the definition and implementation of brain–computer interface (BCI) system relying on brain connectivity (BC) and machine learning/deep learning (DL) for motor imagery (MI)-based applications. During the past few years, many approaches have been explored in terms of types of neurological sources of information, feature extraction, and intention prediction for BCI applications. Two novel aspects are becoming increasingly interesting for the BCI community: BC modeling and DL. The former aims at describing the interactions among different brain regions as connectivity patterns that reflect the dynamics of information flow either at rest or when performing a task. The latter is becoming pervasive for its capability of modeling and predicting complex data, where a huge amount of information is involved. In this scenario, we conducted a systematic literature review on BCI studies that led to the selection of 34 articles meeting all the required criteria. This provides evidence of the rapid growth of the topic over the past few years, though being still in its infancy. The last part of this article is dedicated to this new frontier of BCI that we call MI BC-based computer interfaces highlighting the potential of BC features. This, jointly with DL as enabling technology, has the potential of improving the performance of electroencephalography-based systems.

Exploring the effective signal features of electroencephalogram (EEG) signals is an important issue in the research of brain–computer interface (BCI), and the results can reveal the motor intentions that trigger electrical changes in the brain, which has broad research prospects for feature extraction from EEG data. In contrast to previous EEG decoding methods that are based solely on a convolutional neural network, the traditional convolutional classification algorithm is optimized by combining a transformer mechanism with a constructed end-to-end EEG signal decoding algorithm based on swarm intelligence theory and virtual adversarial training. The use of a self-attention mechanism is studied to expand the receptive field of EEG signals to global dependence and train the neural network by optimizing the global parameters in the model. The proposed model is evaluated on a real-world public dataset and achieves the highest average accuracy of 63.56% in cross-subject experiments, which is significantly higher than that found for recently published algorithms. Additionally, good performance is achieved in decoding motor intentions. The experimental results show that the proposed classification framework promotes the global connection and optimization of EEG signals, which can be further applied to other BCI tasks.

Objective: Most of effectiveness assessments of the widely-used Massage therapy were based on subjective routine clinical assessment tools, such as Visual Analogue Scale (VAS) score. However, few studies demonstrated the impact of massage on the Electroencephalograph (EEG) rhythm decoding of Motor imagery (MI) and motion execution (ME) with trunk left/right bending in patients with skeletal muscle pain. Method: We used the sample entropy (SampEn), permutation entropy (PermuEn), common spatial pattern (CSP) features, support vector machine (SVM) and logic regression (LR) classifiers. We also used the convolutional neural network (CNN) and attention-based bi-directional long short-term memory (BiLSTM) for classification. Results: The averaged SampEn and PermuEn values of alpha rhythm decreased in almost fourteen channels for five statuses (quiet, MI with left/right bending, ME with left/right bending). It indicated that massage alleviates the pain for the patients of skeletal pain. Furthermore, compared with the SVM and LR classifiers, the BiLSTM method achieved a better area under curve (AUC) of 0.89 for the classification of MI with trunk left/right bending before massage. The AUC became smaller after massage than that before massage for the classification of MI with trunk left/right bending using CNN and BiLSTM methods. The Permutation direct indicator (PDI) score showed the significant difference for patients in different statuses (before vs after massage, and MI vs ME). Conclusions: Massage not only affects the quiet status, but also affects the MI and ME. Clinical Impact: Massage therapy may affect a bit on the accuracy of MI with trunk left/right bending and it change the topography of MI and ME with trunk left/right bending for the patients with skeletal muscle pain.

We proposed a classification technique of EEG motor imagery signals using Logistic regression and feature extraction algorithm using filter bank common spatial pattern (FBCSP). Main theme of FBCSP is that the signals decomposed into 5 sub band then calculated CSP for each sub band, this algorithm also allows automated frequency band selection. We combined each subband CSP feature vector, feed this feature vector into machine learning algorithm. In the paper Logistic regression is used to classify among multiple classes. To evaluate this method, we used here publicly available dataset namely Brain-Computer Interface competition IV-2a. Because of high accuracy and kappa that shown in accuracy table that proposed method is promising.

A brain-computer interface (BCI) provides a direct communication pathway between user and external devices. Motor imagery (MI) paradigm is widely used in non-invasive BCI to control external devices by decoding user intentions. The traditional MI-BCI problem is to obtain enough EEG data samples for adopting deep learning techniques, as electroencephalography (EEG) data have intricate and non-stationary properties that can cause a discrepancy between different sessions of data. Because of the discrepancy, the recorded EEG data with different sessions cannot be treated as the same. In this study, we recorded a large intuitive EEG dataset that contained nine types of movements of a single-arm across 12 subjects. We proposed a SessionNet that learns generality with EEG data recorded over multiple sessions using feature similarity to improve classification performance. Additionally, the SessionNet adopts the principle of a hierarchical convolutional neural network that shows robust classification performance regardless of the number of classes. The SessionNet outperforms conventional methods on 3-class, 5-class, and two types of 7-class and 9-class of a single-arm task. Hence, our approach could demonstrate the possibility of using feature similarity based on a novel ensemble learning method to train generality from multiple session data for better MI classification performance.

Motor imagery (MI) is a typical BCI paradigm and has been widely applied into many aspects (e.g. brain-driven wheelchair and motor function rehabilitation training). Although significant achievements have been achieved, multiple motor imagery decoding is still unsatisfactory. To deal with this challenging issue, firstly, a segment of electroencephalogram was extracted and preprocessed. Secondly, we applied a filter bank common spatial pattern (FBCSP) with one-vs-rest (OVR) strategy to extract the spatio-temporal-frequency features of multiple MI. Thirdly, the F-score was employed to optimise and select these features. Finally, the optimized features were fed to the spiking neural networks (SNN) for classification. Evaluation was conducted on two public multiple MI datasets (Dataset IIIa of the BCI competition III and Dataset IIa of the BCI competition IV). Experimental results showed that the average accuracy of the proposed framework reached up to 90.09% (kappa: 0.868) and 81.33% (kappa: 0.751) on the two public datasets, respectively. The achieved performance (accuracy and kappa) was comparable to the best one of the compared methods. This study demonstrated that the proposed method can be used as an alternative approach for multiple MI decoding and it provided a potential solution for online multiple MI detection.

Achieving a reliable classification of motor imagery (MI) tasks is a major challenge in brain-computer interface (BCI) implementation. The set of relevant and discriminative features plays an important role in the classification scheme. This paper presents a supervised approach to select discriminative features for the enhancement of MI classification using multichannel electroencephalography (EEG) signal. The dimension of multiband feature space is reduced using the feature selection method. Each trial of the multichannel EEG signal representing MI tasks is decomposed into a finite set of narrowband signals. The common spatial pattern-based features are extracted from each subband. The features obtained from the multiple subbands are combined to derive a high-dimensional feature vector. The neighborhood component analysis-based feature selection method is implemented to select the features that are relevant in performing an accurate classification. It is a nearest-neighbor-based approach to learn the feature weights with regularization by maximizing the average leave-one-out classification accuracy over the labeled training data. The selected features are used to train the support vector machine for classification. The features relatively irrelevant to the classification task are discarded, yielding a reduction of feature dimension. The evaluation of the proposed method is performed using BCI Competition III dataset 4a and IV dataset 2b. Both are publicly available datasets and are used as types of benchmark data to evaluate the MI classification algorithm to implement BCI. The obtained simulation results confirm the superiority of the proposed method compared to the recently developed algorithms.

Recognition accuracy and response time are both critically essential ahead of building the practical electroencephalography (EEG)-based brain–computer interface (BCI). However, recent approaches have compromised either the classification accuracy or the responding time. This paper presents a novel deep learning approach designed toward both remarkably accurate and responsive motor imagery (MI) recognition based on scalp EEG. Bidirectional long short-term memory (BiLSTM) with the attention mechanism is employed, and the graph convolutional neural network (GCN) promotes the decoding performance by cooperating with the topological structure of features, which are estimated from the overall data. Particularly, this method is trained and tested on the short EEG recording with only 0.4 s in length, and the result has shown effective and efficient prediction based on individual and groupwise training, with 98.81% and 94.64% accuracy, respectively, which outperformed all the state-of-the-art studies. The introduced deep feature mining approach can precisely recognize human motion intents from raw and almost-instant EEG signals, which paves the road to translate the EEG-based MI recognition to practical BCI systems.

The activities for motor imagery (MI) movements in Electroencephalography (EEG) are still interesting and challenging. BCI (Brain Computer Interface) allows the brain signals to control the external devices and also helps a disabled person suffering from neuromuscular disorders. In any BCI system, the two most essential steps are feature extraction and classification method. However, in this paper, the MI classification is improved by the performance of Deep Learning (DL) concept. In this proposed system two-moment imagination of right hand and right foot from the BCI competition three datasets IVA has been taken and classification methods utilizing Conventional neural network (CNN) and Generative Adversarial Network (GAN) are developed. The training time is reduced and non-stationary problem is managed by applying Empirical mode decomposition (EMD) and mixing their intrinsic mode functions (IMFs) in feature extraction technique. The experimental result indicates the proposed GAN classification technique achieves better classification accuracy in terms of 95.29% than the CNN of 89.38%. The proposed GAN method achieves an average positive rate of 62% and average false positive rate of 3.4% on BCI competition three datasets IVA whose EEG facts were resulting from the similar C3, C4, and Cz channels of the motor cortex.

Automatic high-level feature extraction has become a possibility with the advancement of deep learning, and it has been used to optimize efficiency. Recently, classification methods for Convolutional Neural Network (CNN)-based electroencephalography (EEG) motor imagery have been proposed, and have achieved reasonably high classification accuracy. These approaches, however, use the CNN single convolution scale, whereas the best convolution scale varies from subject to subject. This limits the precision of classification. This paper proposes multibranch CNN models to address this issue by effectively extracting the spatial and temporal features from raw EEG data, where the branches correspond to different filter kernel sizes. The proposed method’s promising performance is demonstrated by experimental results on two public datasets, the BCI Competition IV 2a dataset and the High Gamma Dataset (HGD). The results of the technique show a 9.61% improvement in the classification accuracy of multibranch EEGNet (MBEEGNet) from the fixed one-branch EEGNet model, and 2.95% from the variable EEGNet model. In addition, the multibranch ShallowConvNet (MBShallowConvNet) improved the accuracy of a single-scale network by 6.84%. The proposed models outperformed other state-of-the-art EEG motor imagery classification methods.

In brain–computer interface (BCI) systems, motor imagery electroencephalography (MI-EEG) signals are commonly used to detect participant intent. Many factors, including low signal-to-noise ratios and few high-quality samples, make MI classification difficult. In order for BCI systems to function, MI-EEG signals must be studied. In pattern recognition and other fields, deep learning approaches have recently been successfully applied. In contrast, few effective deep learning algorithms have been applied to BCI systems, especially MI-based systems. In this paper, we address these problems from two aspects based on the characteristics of EEG signals: first, we proposed a combined time–frequency domain data enhancement method. This method guarantees that the size of the training data is effectively increased while maintaining the intrinsic composition of the data. Second, our design consists of a parallel CNN that takes both raw EEG images and images transformed through continuous wavelet transform (CWT) as inputs. We conducted classification experiments on a public data set to verify the effectiveness of the algorithm. According to experimental results based on the BCI Competition IV Dataset2a, the average classification accuracy is 97.61%. A comparison of the proposed algorithm with other algorithms shows that it performs better in classification. The algorithm can be used to improve the classification performance of MI-based BCIs and BCI systems created for people with disabilities.

International audience; Motor Imagery Electroencephalogram (MI-EEG) signals are widely used in Brain-Computer Interfaces (BCI). MI-EEG signals of large limbs movements have been explored in recent researches because they deliver relevant classification rates for BCI systems. However, smaller and noisy signals corresponding to hand-finger imagined movements are less frequently used because they are difficult to classify. This study proposes a method for decoding finger imagined movements of the right hand. For this purpose, MI-EEG signals from C3, Cz, P3, and Pz sensors were carefully selected to be processed in the proposed framework. Therefore, a method based on Empirical Mode Decomposition (EMD) is used to tackle the problem of noisy signals. At the same time, the sequence classification is performed by a stacked Bidirectional Long Short-Term Memory (BiLSTM) network. The proposed method was evaluated using k-fold cross-validation on a public dataset, obtaining an accuracy of 82.26%.

When a person attempts intended finger movements, unintended finger movement also occur, a phenomenon called “enslaving”. Given that motor imagery (MI) and motor execution (ME) share a common neural foundation, we hypothesized that the enslaving effect on the spinal motor neuron excitability occurs during MI. To investigate this hypothesis, electromyography (EMG) and F-wave analysis were conducted in 11 healthy male volunteers. Initially, the EMG activity of the left abductor digiti minimi (ADM) muscle during isometric opposition pinch movement by the left thumb and index finger at 50% maximal effort was compared with EMG activity during the Rest condition. Next, the F-wave and background EMG recordings were performed under the Rest condition, followed by the MI condition. Specifically, in the Rest condition, subjects maintained relaxation. In the MI condition, they imagined isometric left thenar muscle activity at 50% maximal voluntary contraction (MVC). During ME, ADM muscle activity was confirmed. During the MI condition, both F-wave persistence and the F-wave/M-wave amplitude ratio obtained from the ADM muscle were significantly increased compared with that obtained during the Rest condition. No difference was observed in the background EMG between the Rest and MI conditions. These results suggest that MI of isometric intended finger muscle activity at 50% MVC facilitates spinal motor neuron excitability corresponding to unintended finger muscle. Furthermore, MI may induce similar modulation of spinal motor neuron excitability as actual movement.

Severely motor-disabled patients, such as those suffering from the so-called “locked-in” syndrome, cannot communicate naturally. They may benefit from brain-computer interfaces (BCIs) exploiting brain signals for communication and therewith circumventing the muscular system. One BCI technique that has gained attention recently is functional near-infrared spectroscopy (fNIRS). Typically, fNIRS-based BCIs allow for brain-based communication via voluntarily modulation of brain activity through mental task performance guided by visual or auditory instructions. While the development of fNIRS-BCIs has made great progress, the reliability of fNIRS-BCIs across time and environments has rarely been assessed. In the present fNIRS-BCI study, we tested six healthy participants across three consecutive days using a straightforward four-choice fNIRS-BCI communication paradigm that allows answer encoding based on instructions using various sensory modalities. To encode an answer, participants performed a motor imagery task (mental drawing) in one out of four time periods. Answer encoding was guided by either the visual, auditory, or tactile sensory modality. Two participants were tested outside the laboratory in a cafeteria. Answers were decoded from the time course of the most-informative fNIRS channel-by-chromophore combination. Across the three testing days, we obtained mean single- and multi-trial (joint analysis of four consecutive trials) accuracies of 62.5 and 85.19%, respectively. Obtained multi-trial accuracies were 86.11% for visual, 80.56% for auditory, and 88.89% for tactile sensory encoding. The two participants that used the fNIRS-BCI in a cafeteria obtained the best single- (72.22 and 77.78%) and multi-trial accuracies (100 and 94.44%). Communication was reliable over the three recording sessions with multi-trial accuracies of 86.11% on day 1, 86.11% on day 2, and 83.33% on day 3. To gauge the trade-off between number of optodes and decoding accuracy, averaging across two and three promising fNIRS channels was compared to the one-channel approach. Multi-trial accuracy increased from 85.19% (one-channel approach) to 91.67% (two-/three-channel approach). In sum, the presented fNIRS-BCI yielded robust decoding results using three alternative sensory encoding modalities. Further, fNIRS-BCI communication was stable over the course of three consecutive days, even in a natural (social) environment. Therewith, the developed fNIRS-BCI demonstrated high flexibility, reliability and robustness, crucial requirements for future clinical applicability.

The classification of electroencephalogram (EEG) signals is of significant importance in brain computer interface (BCI) systems. Aiming to achieve intelligent classification of motor imagery EEG types with high accuracy, a classification methodology using the wavelet packet decomposition and the proposed deep residual convolutional networks (DRes-CNN) is proposed. Firstly, EEG waveforms are segmented into sub-signals. Then the EEG signal features are obtained through the wavelet packet decomposition algorithm, and some selected wavelet coefficients are retained and reconstructed into EEG signals in their respective frequency bands. Subsequently, the reconstructed EEG signals were utilized as input of the proposed deep residual convolutional networks to classify EEG signals. Finally, EEG types of motor imagination are classified by the DRes-CNN classifier intelligently. The datasets from BCI Competition were used to test the performance of the proposed deep learning classifier. Classification experiments show that the average recognition accuracy of this method reaches 98.76%. The proposed method can be further applied to the BCI system of motor imagination control.

Introduction: Advantageous effects of biological motion (BM) detection, a low-perceptual mechanism that allows the rapid recognition and understanding of spatiotemporal characteristics of movement via salient kinematics information, can be amplified when combined with motor imagery (MI), i.e., the mental simulation of motor acts. According to Jeannerod’s neurostimulation theory, asynchronous firing and reduction of mu and beta rhythm oscillations, referred to as suppression over the sensorimotor area, are sensitive to both MI and action observation (AO) of BM. Yet, not many studies investigated the use of BM stimuli using combined AO-MI tasks. In this study, we assessed the neural response in the form of event-related synchronization and desynchronization (ERD/S) patterns following the observation of point-light-walkers and concordant MI, as compared to MI alone.Methods: Twenty right-handed healthy participants accomplished the experimental task by observing BM stimuli and subsequently performing the same movement using kinesthetic MI (walking, cycling, and jumping conditions). We recorded an electroencephalogram (EEG) with 32 channels and performed time-frequency analysis on alpha (8–13 Hz) and beta (18–24 Hz) frequency bands during the MI task. A two-way repeated-measures ANOVA was performed to test statistical significance among conditions and electrodes of interest.Results: The results revealed significant ERD/S patterns in the alpha frequency band between conditions and electrode positions. Post hoc comparisons showed significant differences between condition 1 (walking) and condition 3 (jumping) over the left primary motor cortex. For the beta band, a significantly less difference in ERD patterns (p < 0.01) was detected only between condition 3 (jumping) and condition 4 (reference).Discussion: Our results confirmed that the observation of BM combined with MI elicits a neural suppression, although just in the case of jumping. This is in line with previous findings of AO and MI (AOMI) eliciting a neural suppression for simulated whole-body movements. In the last years, increasing evidence started to support the integration of AOMI training as an adjuvant neurorehabilitation tool in Parkinson’s disease (PD).Conclusion: We concluded that using BM stimuli in AOMI training could be promising, as it promotes attention to kinematic features and imitative motor learning.

Motor Imagery BCI systems have a high rate of users that are not capable of modulating their brain activity accurately enough to communicate with the system. Several studies have identified psychological, cognitive, and neurophysiological measures that might explain this MI-BCI inefficiency. Traditional research had focused on mu suppression in the sensorimotor area in order to classify imagery, but this does not reflect the true dynamics that underlie motor imagery. Functional connectivity reflects the interaction between brain regions during the MI task and resting-state network and is a promising tool in improving MI-BCI classification. In this study, 54 novice MI-BCI users were split into two groups based on their accuracy and their functional connectivity was compared in three network scales (Global, Large and Local scale) during the resting-state, left vs. right-hand motor imagery task, and the transition between the two phases. Our comparison of High and Low BCI performers showed that in the alpha band, functional connectivity in the right hemisphere was increased in High compared to Low aptitude MI-BCI users during motor imagery. These findings contribute to the existing literature that indeed connectivity might be a valuable feature in MI-BCI classification and in solving the MI-BCI inefficiency problem.

Brain-computer interfaces (BCI) have been used to control the gait of a virtual self-avatar with the aim of being used in gait rehabilitation. A BCI decodes the brain signals representing a desire to do something and transforms them into a control command for controlling external devices. The feelings described by the participants when they control a self-avatar in an immersive virtual environment (VE) demonstrate that humans can be embodied in the surrogate body of an avatar (ownership illusion). It has recently been shown that inducing the ownership illusion and then manipulating the movements of one’s self-avatar can lead to compensatory motor control strategies. In order to maximize this effect, there is a need for a method that measures and monitors embodiment levels of participants immersed in virtual reality (VR) to induce and maintain a strong ownership illusion. This is particularly true given that reaching a high level of both BCI performance and embodiment are inter-connected. To reach one of them, the second must be reached as well. Some limitations of many existing systems hinder their adoption for neurorehabilitation: 1- some use motor imagery (MI) of movements other than gait; 2- most systems allow the user to take single steps or to walk but do not allow both, which prevents users from progressing from steps to gait; 3- most of them function in a single BCI mode (cue-paced or self-paced), which prevents users from progressing from machine-dependent to machine-independent walking. Overcoming the aforementioned limitations can be done by combining different control modes and options in one single system. However, this would have a negative impact on BCI performance, therefore diminishing its usefulness as a potential rehabilitation tool. In this case, there will be a need to enhance BCI performance. For such purpose, many techniques have been used in the literature, such as providing modified feedback (whereby the presented feedback is not consistent with the user’s MI), sequential training (recalibrating the classifier as more data becomes available). This thesis was developed over 3 studies. The objective in study 1 was to investigate the possibility of measuring the level of embodiment of an immersive self-avatar, during the performing, observing and imagining of gait, using electroencephalogram (EEG) techniques, by presenting visual feedback that conflicts with the desired movement of embodied participants. The objective of study 2 was to develop and validate a BCI to control single steps and forward walking of an immersive virtual reality (VR) self-avatar, using mental imagery of these actions, in cue-paced and self-paced modes. Different performance enhancement strategies were implemented to increase BCI performance. The data of these two studies were then used in study 3 to construct a generic classifier that could eliminate offline calibration for future users and shorten training time. Twenty different healthy participants took part in studies 1 and 2. In study 1, participants wore an EEG cap and motion capture markers, with an avatar displayed in a head-mounted display (HMD) from a first-person perspective (1PP). They were cued to either perform, watch or imagine a single step forward or to initiate walking on a treadmill. For some of the trials, the avatar took a step with the contralateral limb or stopped walking before the participant stopped (modified feedback). In study 2, participants completed a 4-day sequential training to control the gait of an avatar in both BCI modes. In cue-paced mode, they were cued to imagine a single step forward, using their right or left foot, or to walk forward. In the self-paced mode, they were instructed to reach a target using the MI of multiple steps (switch control mode) or maintaining the MI of forward walking (continuous control mode). The avatar moved as a response to two calibrated regularized linear discriminant analysis (RLDA) classifiers that used the μ power spectral density (PSD) over the foot area of the motor cortex as features. The classifiers were retrained after every session. During the training, and for some of the trials, positive modified feedback was presented to half of the participants, where the avatar moved correctly regardless of the participant’s real performance. In both studies, the participants’ subjective experience was analyzed using a questionnaire. Results of study 1 show that subjective levels of embodiment correlate strongly with the power differences of the event-related synchronization (ERS) within the μ frequency band, and over the motor and pre-motor cortices between the modified and regular feedback trials. Results of study 2 show that all participants were able to operate the cued-paced BCI and the selfpaced BCI in both modes. For the cue-paced BCI, the average offline performance (classification rate) on day 1 was 67±6.1% and 86±6.1% on day 3, showing that the recalibration of the classifiers enhanced the offline performance of the BCI (p < 0.01). The average online performance was 85.9±8.4% for the modified feedback group (77-97%) versus 75% for the non-modified feedback group. For self-paced BCI, the average performance was 83% at switch control and 92% at continuous control mode, with a maximum of 12 seconds of control. Modified feedback enhanced BCI performances (p =0.001). Finally, results of study 3 show that the constructed generic models performed as well as models obtained from participant-specific offline data. The results show that there it is possible to design a participant-independent zero-training BCI., Les interfaces cerveau-ordinateur (ICO) ont été utilisées pour contrôler la marche d'un égo-avatar virtuel dans le but d'être utilisées dans la réadaptation de la marche. Une ICO décode les signaux du cerveau représentant un désir de faire produire un mouvement et les transforme en une commande de contrôle pour contrôler des appareils externes. Les sentiments décrits par les participants lorsqu'ils contrôlent un égo-avatar dans un environnement virtuel immersif démontrent que les humains peuvent être incarnés dans un corps d'un avatar (illusion de propriété). Il a été récemment démontré que provoquer l’illusion de propriété puis manipuler les mouvements de l’égo-avatar peut conduire à des stratégies de contrôle moteur compensatoire. Afin de maximiser cet effet, il existe un besoin d'une méthode qui mesure et surveille les niveaux d’incarnation des participants immergés dans la réalité virtuelle (RV) pour induire et maintenir une forte illusion de propriété. D'autre part, atteindre un niveau élevé de performances (taux de classification) ICO et d’incarnation est interconnecté. Pour atteindre l'un d'eux, le second doit également être atteint. Certaines limitations de plusieurs de ces systèmes entravent leur adoption pour la neuroréhabilitation: 1- certains utilisent l'imagerie motrice (IM) des mouvements autres que la marche; 2- la plupart des systèmes permettent à l'utilisateur de faire des pas simples ou de marcher mais pas les deux, ce qui ne permet pas à un utilisateur de passer des pas à la marche; 3- la plupart fonctionnent en un seul mode d’ICO, rythmé (cue-paced) ou auto-rythmé (self-paced). Surmonter les limitations susmentionnées peut être fait en combinant différents modes et options de commande dans un seul système. Cependant, cela aurait un impact négatif sur les performances de l’ICO, diminuant ainsi son utilité en tant qu'outil potentiel de réhabilitation. Dans ce cas, il sera nécessaire d'améliorer les performances des ICO. À cette fin, de nombreuses techniques ont été utilisées dans la littérature, telles que la rétroaction modifiée, le recalibrage du classificateur et l'utilisation d'un classificateur générique. Le projet de cette thèse a été réalisé en 3 études, avec objectif d'étudier dans l'étude 1, la possibilité de mesurer le niveau d'incarnation d'un égo-avatar immersif, lors de l'exécution, de l'observation et de l'imagination de la marche, à l'aide des techniques encéphalogramme (EEG), en présentant une rétroaction visuelle qui entre en conflit avec la commande du contrôle moteur des sujets incarnés. L'objectif de l'étude 2 était de développer un BCI pour contrôler les pas et la marche vers l’avant d'un égo-avatar dans la réalité virtuelle immersive, en utilisant l'imagerie motrice de ces actions, dans des modes rythmés et auto-rythmés. Différentes stratégies d'amélioration des performances ont été mises en œuvre pour augmenter la performance (taux de classification) de l’ICO. Les données de ces deux études ont ensuite été utilisées dans l'étude 3 pour construire des classificateurs génériques qui pourraient éliminer la calibration hors ligne pour les futurs utilisateurs et raccourcir le temps de formation. Vingt participants sains différents ont participé aux études 1 et 2. Dans l'étude 1, les participants portaient un casque EEG et des marqueurs de capture de mouvement, avec un avatar affiché dans un casque de RV du point de vue de la première personne (1PP). Ils ont été invités à performer, à regarder ou à imaginer un seul pas en avant ou la marche vers l’avant (pour quelques secondes) sur le tapis roulant. Pour certains essais, l'avatar a fait un pas avec le membre controlatéral ou a arrêté de marcher avant que le participant ne s'arrête (rétroaction modifiée). Dans l'étude 2, les participants ont participé à un entrainement séquentiel de 4 jours pour contrôler la marche d'un avatar dans les deux modes de l’ICO. En mode rythmé, ils ont imaginé un seul pas en avant, en utilisant leur pied droit ou gauche, ou la marche vers l’avant . En mode auto-rythmé, il leur a été demandé d'atteindre une cible en utilisant l'imagerie motrice (IM) de plusieurs pas (mode de contrôle intermittent) ou en maintenir l'IM de marche vers l’avant (mode de contrôle continu). L'avatar s'est déplacé en réponse à deux classificateurs ‘Regularized Linear Discriminant Analysis’ (RLDA) calibrés qui utilisaient comme caractéristiques la densité spectrale de puissance (Power Spectral Density; PSD) des bandes de fréquences µ (8-12 Hz) sur la zone du pied du cortex moteur. Les classificateurs ont été recalibrés après chaque session. Au cours de l’entrainement et pour certains des essais, une rétroaction modifiée positive a été présentée à la moitié des participants, où l'avatar s'est déplacé correctement quelle que soit la performance réelle du participant. Dans les deux études, l'expérience subjective des participants a été analysée à l'aide d'un questionnaire. Les résultats de l'étude 1 montrent que les niveaux subjectifs d’incarnation sont fortement corrélés à la différence de la puissance de la synchronisation liée à l’événement (Event-Related Synchronization; ERS) sur la bande de fréquence μ et sur le cortex moteur et prémoteur entre les essais de rétroaction modifiés et réguliers. L'étude 2 a montré que tous les participants étaient capables d’utiliser le BCI rythmé et auto-rythmé dans les deux modes. Pour le BCI rythmé, la performance hors ligne moyenne au jour 1 était de 67±6,1% et 86±6,1% au jour 3, ce qui montre que le recalibrage des classificateurs a amélioré la performance hors ligne du BCI (p

In recent years, more and more frameworks have been applied to brain-computer interface technology, and electroencephalogram-based motor imagery (MI-EEG) is developing rapidly. However, it is still a challenge to improve the accuracy of MI-EEG classification. A deep learning framework termed IS-CBAM-convolutional neural network (CNN) is proposed to address the non-stationary nature, the temporal localization of excitation occurrence, and the frequency band distribution characteristics of the MI-EEG signal in this paper. First, according to the logically symmetrical relationship between the C3 and C4 channels, the result of the time-frequency image subtraction (IS) for the MI-EEG signal is used as the input of the classifier. It both reduces the redundancy and increases the feature differences of the input data. Second, the attention module is added to the classifier. A convolutional neural network is built as the base classifier, and information on the temporal location and frequency distribution of MI-EEG signal occurrences are adaptively extracted by introducing the Convolutional Block Attention Module (CBAM). This approach reduces irrelevant noise interference while increasing the robustness of the pattern. The performance of the framework was evaluated on BCI competition IV dataset 2b, where the mean accuracy reached 79.6%, and the average kappa value reached 0.592. The experimental results validate the feasibility of the framework and show the performance improvement of MI-EEG signal classification.

Background Brain wave signal recognition has gained increased attention in neuro-rehabilitation applications. This has driven the development of brain–computer interface (BCI) systems. Brain wave signals are acquired using electroencephalography (EEG) sensors, processed and decoded to identify the category to which the signal belongs. Once the signal category is determined, it can be used to control external devices. However, the success of such a system essentially relies on significant feature extraction and classification algorithms. One of the commonly used feature extraction technique for BCI systems is common spatial pattern (CSP). Results The performance of the proposed spatial-frequency-temporal feature extraction (SPECTRA) predictor is analysed using three public benchmark datasets. Our proposed predictor outperformed other competing methods achieving lowest average error rates of 8.55%, 17.90% and 20.26%, and highest average kappa coefficient values of 0.829, 0.643 and 0.595 for BCI Competition III dataset IVa, BCI Competition IV dataset I and BCI Competition IV dataset IIb, respectively. Conclusions Our proposed SPECTRA predictor effectively finds features that are more separable and shows improvement in brain wave signal recognition that can be instrumental in developing improved real-time BCI systems that are computationally efficient.

Nowadays, motor imagery (MI) electroencephalogram (EEG) signal classification has become a hotspot in the research field of brain computer interface (BCI). More recently, deep learning has emerged as a promising technique to automatically extract features of raw MI EEG signals and then classify them. However, deep learning-based methods still face two challenging problems in practical MI EEG signal classification applications: (1) Generally, training a deep learning model successfully needs a large amount of labeled data. However, most of the EEG signal data is unlabeled and it is quite difficult or even impossible for human experts to label all the signal samples manually. (2) It is extremely time-consuming and computationally expensive to train a deep learning model from scratch. To cope with these two challenges, a deep transfer convolutional neural network (CNN) framework based on VGG-16 is proposed for EEG signal classification. The proposed framework consists of a VGG-16 CNN model pre-trained on the ImageNet and a target CNN model which shares the same structure with VGG-16 except for the softmax output layer. The parameters of the pre-trained VGG-16 CNN model are directly transferred to the target CNN model used for MI EEG signal classification. Then, front-layers parameters in the target model are frozen, while later-layers parameters are fine-tuned by the target MI dataset. The target dataset is composed of time-frequency spectrum images of EEG signals. The performance of the proposed framework is verified on the public benchmark dataset 2b from the BCI competition IV. The experimental results show that the proposed framework improves the accuracy and efficiency performance of EEG signal classification compared with traditional methods, including support vector machine (SVM), artificial neural network (ANN), and standard CNN.

The classification of motor imagery Electroencephalogram (EEG) of the same limb is important for natural control of neuroprosthesis. Due to the close spatial representations on the motor cortex area of the brain, the discrimination of the different motor imagery tasks is challenging. In this paper, phase synchronization information was proposed to classify motor imagery EEG within the same limb. In addition, non-portable was compared with portable EEG acquisition equipment for the purpose of making the brain computer interface (BCI) system more practical. In the non-portable case, the average accuracy of the binary classification and 3-class classification was 60.6% and 42.7%. In the portable case, the average EEG decoding accuracy of 58.5% and 39.9% was achieved for the two and three tasks. Furthermore, in both two cases, different sets of electrode pairs got the similar results. Moreover, we found that the proposed phase information based method was less sensitive to the number of EEG channels and had less performance degradation in portable EEG equipment. These results show it is possible to use phase synchronization information to discriminate different motor imagery tasks within the same limb. Eventually, this will potentially make the control of neuroprosthesis or other rehabilitation device more natural and intuitive.

Feature extraction and classification play an important role in brain–computer interface (BCI) systems. In traditional approaches, methods in pattern recognition field are adopted to solve these problems. Nowadays, the deep learning theory has developed so fast that researchers have employed it in many areas like computer vision and speech recognition, which have achieved remarkable results. However, few people introduce the deep learning method into the study of biomedical signals, especially EEG signals. In this paper, a wavelet transform-based input, which combines the time-frequency images of C3, Cz, and C4 channels, is proposed to extract the feature of motor imagery EEG signal. Then, a 2-Layer convolutional neural network is built as the classifier and convolutional kernels of different sizes are validated. The performance obtained by the proposed approach is evaluated by accuracy and Kappa value. The accuracy on dataset III from BCI competition II reaches 90%, and the best Kappa value on dataset 2a from competition IV is greater than many of other methods. In addition, the proposed method utilizes a resized small input, which reduces calculation complexity, so the training period is relatively faster. The results show that the method using convolutional neural network can be comparable or better than the other state-of-the-art approaches, and the performance will be improved when there is sufficient data.

Objective. Processing strategies are analyzed with respect to the classification of electroencephalographic signals related to brain-computer interfaces (BCIs) based on motor imagery (MI). A review of literature is carried out to understand the achievements in MI classification, the most promising trends, and the challenges in replicating these results. Main focus is placed on performance by means of a rigorous metrological analysis carried out in compliance with the international vocabulary of metrology. Hence, classification accuracy and its uncertainty are considered, as well as repeatability and reproducibility. Approach. The paper works included in the review concern the classification of electroencephalographic signals in motor-imagery-based BCIs. Article search was carried out in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses standard and 89 studies were included. Main results. Statistically-based analyses show that brain-inspired approaches are increasingly proposed, and that these are particularly successful in discriminating against multiple classes. Notably, many proposals involve convolutional neural networks. Instead, classical machine learning approaches are still effective for binary classifications. Many proposals combine common spatial pattern, least absolute shrinkage and selection operator, and support vector machines. Regarding reported classification accuracies, performance above the upper quartile is in the 85%–100% range for the binary case and in the 83%–93% range for multi-class one. Associated uncertainties are up to 6% while repeatability for a predetermined dataset is up to 8%. Reproducibility assessment was instead prevented by lack of standardization in experiments. Significance. By relying on the analyzed studies, the reader is guided towards the development of a successful processing strategy as a crucial part of a BCI. Moreover, it is suggested that future studies should extend these approaches on data from more subjects and with custom experiments, even by investigating online operation. This would also enable the quantification of the results reproducibility.

Motor imagery (MI) is a classical method of brain–computer interaction (BCI), in which electroencephalogram (EEG) signal features evoked by imaginary body movements are recognized, and relevant information is extracted. Recently, various deep-learning methods are being focused on in finding an easy-to-use EEG representation method that can preserve both temporal information and spatial information. To further utilize the spatial and temporal features of EEG signals, an improved 3D representation of the EEG and a densely connected multi-branch 3D convolutional neural network (dense M3D CNN) for MI classification are introduced in this paper. Specifically, as compared to the original 3D representation, a new padding method is proposed to pad the points without electrodes with the mean of all the EEG signals. Based on this new 3D presentation, a densely connected multi-branch 3D CNN with a novel dense connectivity is proposed for extracting the EEG signal features. Experiments were carried out on the WAY-EEG-GAL and BCI competition IV 2a datasets to verify the performance of this proposed method. The experimental results show that the proposed framework achieves a state-of-the-art performance that significantly outperforms the multi-branch 3D CNN framework, with a 6.208% improvement in the average accuracy for the BCI competition IV 2a datasets and 6.281% improvement in the average accuracy for the WAY-EEG-GAL datasets, with a smaller standard deviation. The results also prove the effectiveness and robustness of the method, along with validating its use in MI-classification tasks.

Brain Computer Interface technologies are popular methods of communication between the human brain and external devices. One of the most popular approaches to BCI is Motor Imagery. In BCI applications, the ElectroEncephaloGraphy is a very popular measurement for brain dynamics because of its non-invasive nature. Although there is a high interest in the BCI topic, the performance of existing systems is still far from ideal, due to the difficulty of performing pattern recognition tasks in EEG signals. BCI systems are composed of a wide range of components that perform signal pre-processing, feature extraction and decision making. In this paper, we define a BCI Framework, named Enhanced Fusion Framework, where we propose three different ideas to improve the existing MI-based BCI frameworks. Firstly, we include aan additional pre-processing step of the signal: a differentiation of the EEG signal that makes it time-invariant. Secondly, we add an additional frequency band as feature for the system and we show its effect on the performance of the system. Finally, we make a profound study of how to make the final decision in the system. We propose the usage of both up to six types of different classifiers and a wide range of aggregation functions (including classical aggregations, Choquet and Sugeno integrals and their extensions and overlap functions) to fuse the information given by the considered classifiers. We have tested this new system on a dataset of 20 volunteers performing motor imagery-based brain-computer interface experiments. On this dataset, the new system achieved a 88.80% of accuracy. We also propose an optimized version of our system that is able to obtain up to 90,76%. Furthermore, we find that the pair Choquet/Sugeno integrals and overlap functions are the ones providing the best results., IEEE Transactions on Cybernetics (2021)

[ES] Las interfaces cerebro-computador pueden ser de gran ayuda para aquellas personas que sufren de alguna discapacidad motora severa, permitiendo así mejorar su calidad de vida. Por ello, este trabajo se basa en la realización de una comparación entre diferentes paradigmas utilizados en las interfaces cerebro-computador, los cuales en este caso son P300, Steady-State Visual Evoked Potentials y Motor Imagery. Asimismo, el objetivo principal del trabajo se basa en la implementación de un speller con el paradigma que mejores resultados obtenga, tanto a nivel de usuario como de rendimiento, tras la realizar la comparación entre ellos. Por último, la eficacia del speller se va a comprobar mediante la ejecución de una serie de pruebas en distintos usuarios, para así observar su funcionamiento independientemente del sujeto que se encuentre utilizándolo., [EN] Brain-computer interfaces can be of great help for those who suffer from a severe motor disability, thus improving their quality of life. Therefore, this work is based on a comparison between different paradigms used in brain-computer interfaces, which in this case are P300, Steady-State Visual Evoked Potentials and Motor Imagery. Likewise, the main objective of the work is based on the implementation of a speller with the paradigm that obtains the best results, both at the user and performance level, after comparing them. Finally, the effectiveness of the speller will be verified by running a series of tests on different users, in order to observe its operation regardless of the subject who is using it.

With the development of technology and the rise of the meta-universe concept, the brain-computer interface (BCI) has become a hotspot in the research field, and the BCI based on motor imagery (MI) EEG has been widely concerned. However, in the process of MI-EEG decoding, the performance of the decoding model needs to be improved. At present, most MI-EEG decoding methods based on deep learning cannot make full use of the temporal and frequency features of EEG data, which leads to a low accuracy of MI-EEG decoding. To address this issue, this paper proposes a two-branch convolutional neural network (TBTF-CNN) that can simultaneously learn the temporal and frequency features of EEG data. The structure of EEG data is reconstructed to simplify the spatio-temporal convolution process of CNN, and continuous wavelet transform is used to express the time-frequency features of EEG data. TBTF-CNN fuses the features learned from the two branches and then inputs them into the classifier to decode the MI-EEG. The experimental results on the BCI competition IV 2b dataset show that the proposed model achieves an average classification accuracy of 81.3% and a kappa value of 0.63. Compared with other methods, TBTF-CNN achieves a better performance in MI-EEG decoding. The proposed method can make full use of the temporal and frequency features of EEG data and can improve the decoding accuracy of MI-EEG.

Motor deficiencies constitute a significant problem affecting millions of people worldwide. Such people suffer from a debility in daily functioning, which may lead to decreased and incoherence in daily routines and deteriorate their quality of life (QoL). Thus, there is an essential need for assistive systems to help those people achieve their daily actions and enhance their overall QoL. This study proposes a novel brain&ndash, computer interface (BCI) system for assisting people with limb motor disabilities in performing their daily life activities by using their brain signals to control assistive devices. The extraction of useful features is vital for an efficient BCI system. Therefore, the proposed system consists of a hybrid feature set that feeds into three machine-learning (ML) classifiers to classify motor Imagery (MI) tasks. This hybrid feature selection (FS) system is practical, real-time, and an efficient BCI with low computation cost. We investigate different combinations of channels to select the combination that has the highest impact on performance. The results indicate that the highest achieved accuracies using a support vector machine (SVM) classifier are 93.46% and 86.0% for the BCI competition III&ndash, IVa dataset and the autocalibration and recurrent adaptation dataset, respectively. These datasets are used to test the performance of the proposed BCI. Also, we verify the effectiveness of the proposed BCI by comparing its performance with recent studies. We show that the proposed system is accurate and efficient. Future work can apply the proposed system to individuals with limb motor disabilities to assist them and test their capability to improve their QoL. Moreover, the forthcoming work can examine the system&rsquo, s performance in controlling assistive devices such as wheelchairs or artificial limbs.

Background: advanced paraclinical approaches using functional neuroimaging and electroencephalography (EEG) allow identifying patients who are covertly aware despite being diagnosed as unresponsive wakefulness syndrome (UWS). Bedside detection of covert awareness employing motor imagery tasks (MI), which is a universally accepted clinical indicator of awareness in the absence of overt behavior, may miss some of these patients, as they could still have a certain level of awareness. We aimed at assessing covert awareness in patients with UWS using a visuomotor-guided motor imagery task (VMI) during EEG recording. Methods: nine patients in a minimally conscious state (MCS), 11 patients in a UWS, and 15 healthy individuals (control group—CG) were provided with an VMI (imagine dancing while watching a group dance video to command), a simple-MI (imagine squeezing their right hand to command), and an advanced-MI (imagine dancing without watching a group dance video to command) to detect command-following. We analyzed the command-specific EEG responses (event-related synchronization/desynchronization—ERS/ERD) of each patient, assessing whether these responses were appropriate, consistent, and statistically similar to those elicited in the CG, as reliable markers of motor imagery. Results: All patients in MCS, all healthy individuals and one patient in UWS repeatedly and reliably generated appropriate EEG responses to distinct commands of motor imagery with a classification accuracy of 60–80%. Conclusions: VMI outperformed significantly MI tasks. Therefore, patients in UWS may be still misdiagnosed despite a rigorous clinical assessment and an appropriate MI assessment. It is thus possible to suggest that motor imagery tasks should be delivered to patients with chronic disorders of consciousness in visuomotor-aided modality (also in the rehabilitation setting) to greatly entrain patient’s participation. In this regard, the EEG approach we described has the clear advantage of being cheap, portable, widely available, and objective. It may be thus considered as, at least, a screening tool to identify the patients who deserve further, advanced paraclinical approaches.

A brain-computer interface (BCI) based on electroencephalography (EEG) can provide independent information exchange and control channels for the brain and the outside world. However, EEG signals come from multiple electrodes, the data of which can generate multiple features. How to select electrodes and features to improve classification performance has become an urgent problem to be solved. This paper proposes a deep convolutional neural network (CNN) structure with separated temporal and spatial filters, which selects the raw EEG signals of the electrode pairs over the motor cortex region as hybrid samples without any preprocessing or artificial feature extraction operations. In the proposed structure, a 5-layer CNN has been applied to learn EEG features, a 4-layer max pooling has been used to reduce dimensionality, and a fully-connected (FC) layer has been utilized for classification. Dropout and batch normalization are also employed to reduce the risk of overfitting. In the experiment, the 4 s EEG data of 10, 20, 60, and 100 subjects from the Physionet database are used as the data source, and the motor imaginations (MI) tasks are divided into four types: left fist, right fist, both fists, and both feet. The results indicate that the global averaged accuracy on group-level classification can reach 97.28%, the area under the receiver operating characteristic (ROC) curve stands out at 0.997, and the electrode pair with the highest accuracy on 10 subjects dataset is FC3-FC4, with 98.61%. The research results also show that this CNN classification method with minimal (2) electrode can obtain high accuracy, which is an advantage over other methods on the same database. This proposed approach provides a new idea for simplifying the design of BCI systems, and accelerates the process of clinical application.

Common spatial pattern (CSP) method is widely used for spatial filtering and brain pattern extraction from electroencephalogram (EEG) signals in motor imagery (MI)-based brain-computer interfaces (BCIs). The participant-specific time window relative to the visual cue has a significant impact on the effectiveness of the CSP. However, the time window is usually selected experientially or manually. To solve this problem, we propose a novel feature selection approach for MI-based BCIs. Specifically, multiple time segments were obtained by decomposing each EEG sample of the MI task. Furthermore, the features were extracted by CSP from each time segment and were combined to form a new feature vector. Finally, the optimal temporal combination patterns for the new feature vector were selected based on four feature selection algorithms, i.e., mutual information, least absolute shrinkage and selection operator, principal component analysis and stepwise linear discriminant analysis (denoted as MUIN, LASSO, PCA, and SWLDA, respectively), and the classification algorithm was employed to evaluate the average classification accuracy. With three BCI competition datasets, the results of the four proposed algorithms were compared with traditional CSP algorithm in classification accuracy. Experimental results show that compared with traditional algorithm, the proposed methods significantly improve performance. Specifically, the LASSO achieved the highest accuracy (88.58%) among the proposed methods. Importantly, the average classification accuracies using the proposed approaches significantly improved 10.14% (MUIN), 11.40% (LASSO), 6.08% (PCA), and 10.25% (SWLDA) compared to that using CSP. These results indicate that the proposed approach is expected to be practical in MI-based BCIs.

Strokes are a growing cause of mortality and many stroke survivors suffer from motor impairment as well as other types of disabilities in their daily life activities. To treat these sequelae, motor imagery (MI) based brain-computer interface (BCI) systems have shown potential to serve as an effective neurorehabilitation tool for post-stroke rehabilitation therapy. In this review, different MI-BCI based strategies, including “Functional Electric Stimulation, Robotics Assistance and Hybrid Virtual Reality based Models,” have been comprehensively reported for upper-limb neurorehabilitation. Each of these approaches have been presented to illustrate the in-depth advantages and challenges of the respective BCI systems. Additionally, the current state-of-the-art and main concerns regarding BCI based post-stroke neurorehabilitation devices have also been discussed. Finally, recommendations for future developments have been proposed while discussing the BCI neurorehabilitation systems.

Cooperative interactions among neural groups scattered in adjacent brain regions emerge in cognitive acts, which implies that local spatial relationships between EEG channels could benefit the classification for motor imagery (MI) tasks. Due to the lack of research exploring this issue, this paper propose a novel feature extraction method, termed channel binary pattern (CBP), to extract the local spatial information. CBP is used to discover local spatial patterns by binarizing the interaction of adjacent local regions. Besides, we further propose a new feature extraction method for the global spatial feature termed a combination of the multiclass spatial pattern (CMSP), which computes the maximal global variance for each class. Then, the fusion of local and global information is performed by the proposed algorithm of spatial information fusion based on electroencephalography (EEG-SIF), which is able to explore comprehensive spatial information for brain activity when MI occurs. An experimental study is implemented with BCI Competition IV Dataset 2a. And the superior classification result confirms the effectiveness of spatial information extraction in different scales (local and global space) and the feasibility of information fusion.

The motor imagery-based brain-computer interface (BCI) using electroencephalography (EEG) has been receiving attention from neural engineering researchers and is being applied to various rehabilitation applications. However, the performance degradation caused by motor imagery EEG with very low single-to-noise ratio faces several application issues with the use of a BCI system. In this paper, we propose a novel motor imagery classification scheme based on the continuous wavelet transform and the convolutional neural network. Continuous wavelet transform with three mother wavelets is used to capture a highly informative EEG image by combining time-frequency and electrode location. A convolutional neural network is then designed to both classify motor imagery tasks and reduce computation complexity. The proposed method was validated using two public BCI datasets, BCI competition IV dataset 2b and BCI competition II dataset III. The proposed methods were found to achieve improved classification performance compared with the existing methods, thus showcasing the feasibility of motor imagery BCI.

In recent years, the control of devices “by the power of the mind” has become a very controversial topic but has also been very well researched in the field of state-of-the-art gadgets, such as smartphones, laptops, tablets and even smart TVs, and also in medicine, to be used by people with disabilities for whom these technologies may be the only way to communicate with the outside world. It is well known that BCI control is a skill and can be improved through practice and training. This paper aims to improve and diversify signal processing methods for the implementation of a brain-computer interface (BCI) based on neurological phenomena recorded during motor tasks using motor imagery (MI). The aim of the research is to extract, select and classify the characteristics of electroencephalogram (EEG) signals, which are based on sensorimotor rhythms, for the implementation of BCI systems. This article investigates systems based on brain-computer interfaces, especially those that use the electroencephalogram as a method of acquisition of MI tasks. The purpose of this article is to allow users to manipulate quadcopter virtual structures (external, robotic objects) simply through brain activity, correlated with certain mental tasks using undecimal transformation (UWT) to reduce noise, Independent Component Analysis (ICA) together with determination coefficient (r2) and, for classification, a hybrid neural network consisting of Radial Basis Functions (RBF) and a multilayer perceptron–recurrent network (MLP–RNN), obtaining a classification accuracy of 95.5%. Following the tests performed, it can be stated that the use of biopotentials in human–computer interfaces is a viable method for applications in the field of BCI. The results presented show that BCI training can produce a rapid change in behavioral performance and cognitive properties. If more than one training session is used, the results may be beneficial for increasing poor cognitive performance. To achieve this goal, three steps were taken: understanding the functioning of BCI systems and the neurological phenomena involved; acquiring EEG signals based on sensorimotor rhythms recorded during MI tasks; applying and optimizing extraction methods, selecting and classifying characteristics using neuronal networks.

Producción Científica, Neurofeedback training (NFT) has shown promising results in recent years as a tool to address the effects of age-related cognitive decline in the elderly. Since previous studies have linked reduced complexity of electroencephalography (EEG) signal to the process of cognitive decline, we propose the use of non-linear methods to characterise changes in EEG complexity induced by NFT. In this study, we analyse the pre- and post-training EEG from 11 elderly subjects who performed an NFT based on motor imagery (MI–NFT). Spectral changes were studied using relative power (RP) from classical frequency bands (delta, theta, alpha, and beta), whilst multiscale entropy (MSE) was applied to assess EEG-induced complexity changes. Furthermore, we analysed the subject’s scores from Luria tests performed before and after MI–NFT. We found that MI–NFT induced a power shift towards rapid frequencies, as well as an increase of EEG complexity in all channels, except for C3. These improvements were most evident in frontal channels. Moreover, results from cognitive tests showed significant enhancement in intellectual and memory functions. Therefore, our findings suggest the usefulness of MI–NFT to improve cognitive functions in the elderly and encourage future studies to use MSE as a metric to characterise EEG changes induced by MI–NFT., Ministerio de Ciencia e Innovación (Grants PID2020-115468RB-I00 and RTC2019- 007350-1), Gobierno de España (Agencia Estatal de Investigación) - (Projects 10.13039/ 501100011033), Unión Europea y Fondo Europeo de Desarrollo Regional (FEDER) - (Cooperation Programme Interreg V-A Spain-Portugal POCTEP 2014–2020)

Worldwide, millions of people are locked in or in a wheelchair, due to several neuromuscular disorders or spinal cord injuries. These individuals are deprived of trivial social activities, like interacting or playing games with other people. Such activities are crucial for personal development, and can have a great impact on the quality of their lives. This work aims at the design and implementation of an electroencephalography (EEG) based motor imagery (MI) brain computer interface (BCI) system that would allow disabled, and able-bodied, individuals alike to control a drone in a 3D physical environment by only using their thoughts. An improved version of the filter bank common spatial pattern (FBCSP) algorithm was developed, and it has shown to perform superior (68.5% accuracy) to the winning FBCSP algorithm (67.8% accuracy), when tested on dataset 2a (4 class MI) of the BCI competition IV. A deep convolutional neural network (CNN) based algorithm was also implemented and tested on the same dataset, which however performed inferior (62.9% accuracy) to the winner, as well as our proposed FBCSP algorithms. The improved FBCSP was then tested on our in-house 5-class (left hand, right hand, tongue, both feet and rest) MI dataset (collected from 10 able-bodied subjects) and obtained a mean accuracy of 41.8±11.74%. This is considered a significant result though it is not good enough to attempt the control of a real drone.

An important goal in Brain-Computer Interfacing (BCI) is tofind and enhance procedural strategies for users for whom BCI control is not sufficiently accurate. To address this challenge, we conducted offline analyses and online experiments to test whether the classification of different types of motor imagery could be improved when the training of the classifier was performed on the data obtained with the assistive muscular stimulation below the motor threshold. 10 healthy participants underwent three different types of experimental conditions: a) Motor imagery (MI) of hands and feet b) sensory threshold neuromuscular electrical stimulation (STM) of hands and feet while resting and c) sensory threshold neuromuscular electrical stimulation during performance of motor imagery (BOTH). Also, another group of 10 participants underwent conditions a) and c). Then, online experiments with 15 users were performed. These subjects received neurofeedback during MI using classifiers calibrated either on MIor BOTH data recorded in the same experiment. Offline analyses showed that decoding MI alone using a classifier based on BOTH resulted in a better BCI accuracy compared to using a classifier based on MI alone. Online experiments confirmed accuracy improvement of MI alone being decoded with the classifier trained on BOTH data. In addition, we observed that the performance in MI condition could be predicted on the basis of a more pronounced connectivity within sensorimotor areas in the frequency bands providing the best performance in BOTH. Thesefinding might offer a new avenue for training SMR-based BCI systems particularly for users having difficulties to achieve efficient BCI control. It might also be an alternative strategy for users who cannot perform real movements but still have remaining afferent pathways (e.g., ALS and stroke patients). The work of CV and KRM was funded by the German Ministry for Education and Research (BMBF) under Grant 01IS14013A-E and Grant 01GQ1115, as well as by the Deutsche Forschungsgesellschaft (DFG) under Grant MU 987/19-1, MU987/14-1 and DFG MU 987/3-2 and by the EU-FP7 MUNDUS project Grant 248326. Additionally, KRM was partly supported by the German Ministry for Education and Research as Berlin Big Data Centre (01IS14013A) and Berlin Center for Machine Learning (01IS18037I). Partial funding by DFG is acknowledged (EXC 2046/1, project-ID: 390685689). This work was also supported by the Institute for Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (No. 2017-0-00451, No. 2017-0-01779). CV was also supported by the Spanish Ministry of Economy RYC-2014-15671. VN has been supported by the HSE Basic Research Program, Russian Academic Excellence Project '5-100'. Correspondence to CV, KRM and VN.

This pilot study implements a hybrid BCI system in an effort to deduce the effects of measuring more than one brain signal in a motor imagery (MI) task. In addition to sensorimotor rhythms (SMRs), a steady state visual evoked potential (SSVEP) was introduced to acquire additional information relating to user intention. A common spatial pattern (CSP) filter followed by a support vector machine (SVM) classifier were used to distinguish between MI and the resting state. The power spectral density (PSD) was used to classify the SSVEP. Results from online simulations of EEG data collected from 10 able-bodied participants showed that the hybrid BCI’s performance achieved a classification accuracy of 77.3±8.2%, with an SSVEP classification accuracy of 94.4±3.5%, and MI classification accuracy of 80.9±8.1%, an improvement upon purely MI-based multi-class BCI paradigms.

Despite the growing interest in brain–machine interface (BMI)-driven neuroprostheses, the translation of the BMI output into a suitable control signal for the robotic device is often neglected. In this article, we propose a novel control approach based on dynamical systems that was explicitly designed to take into account the nature of the BMI output that actively supports the user in delivering real-valued commands to the device and, at the same time, reduces the false positive rate. We hypothesize that such a control framework would allow users to continuously drive a mobile robot and it would enhance the navigation performance. 13 healthy users evaluated the system during three experimental sessions. Users exploit a 2-class motor imagery BMI to drive the robot to five targets in two experimental conditions: with a discrete control strategy, traditionally exploited in the BMI field, and with the novel continuous control framework developed herein. Experimental results show that the new approach: 1) allows users to continuously drive the mobile robot via BMI; 2) leads to significant improvements in the navigation performance; and 3) promotes a better coupling between user and robot. These results highlight the importance of designing a suitable control framework to improve the performance and the reliability of BMI-driven neurorobotic devices.

[ES] La interfaz cerebro-ordenador es el camino para conectar es sistema neuromuscular y el mundo externo. El presente TFG consiste en la adquisición, procesado y clasificación de señales de electroencefalograma (EEG) de imaginación de movimientos motores (MI). La imaginación de movimientos motores es un estado cognitivo fruto de la representación interna de una acción motora específica sin actividad muscular. En primer lugar, se adquieren señales de EEG multicanal de 18 sujetos diferentes mientras la imaginación de tareas afectivas y neutrales. La tarea afectiva consiste en imaginarse apretar un cactus con la mano derecha varias veces, sintiendo una emoción desagradable. En la condición neutral, la tarea es imaginarse apretar una pelota normal o una pelota de masaje. Tras la adquisición de las señales de EEG se lleva a cabo el procesamiento de estas con MATLAB y EEGLAB. Para estudiar el incremento o disminución de la actividad neuronal se emplea la Perturbación en el Espectro Relacionada con Eventos (ERSP). Finalmente, se extraen las características que permitan clasificar las señales mediante un Análisis Discriminante Lineal de contracción (sLDA). Esta clasificación se hace separadamente para cada sujeto y para evaluar su resultado se emplea la precisión., [EN] Brain-Computer Interfaces is an alternative pathway to connect the brain with the outside world besides the neuro-muscular system. The current project consists of the acquisition, processing and classification of the signal of the motor imagery electroencephalogram (EEG). Motor imagery is a dynamic cognitive state during which the representation of a specific motor action is internally created without any muscular activity. First, 18 subjects are instructed to imagine affective, and neutral motor tasks while multichannel EEG is recorded simultaneously. In the affective MI condition, they had to imagine squeezing a cactus with their right hand several times, eliciting an unpleasant emotion. In the neutral condition their task was to imagine squeezing a ball. After the acquisition, the signal is processed with MATLAB and EEGLAB. To study the increase or decrease of neural activity Event-related Spectral Perturbation (ERSP) is calculated. Finally, the study and extraction of the features of the signal is done for the posterior Shrinkage Linear Discriminant Analysis (sLDA) classification. The classification is done separately for each subject and, to evaluate it, it is used the accuracy

A human–computer interaction (HCI) system can be used to detect different categories of the brain wave signals that can be beneficial for neurorehabilitation, seizure detection and sleep stage classification. Research on developing HCI systems using brain wave signals has progressed a lot over the years. However, real-time implementation, computational complexity and accuracy are still a concern. In this work, we address the problem of selecting the appropriate filtering frequency band while also achieving a good system performance by proposing a frequency-based approach using long short-term memory network (LSTM) for recognizing different brain wave signals. Adaptive filtering using genetic algorithm is incorporated for a hybrid system utilizing common spatial pattern and LSTM network. The proposed method (OPTICAL+) achieved an overall average classification error rate of 30.41% and a kappa coefficient value of 0.398, outperforming the state-of-the-art methods. The proposed OPTICAL+ predictor can be used to develop improved HCI systems that will aid in neurorehabilitation and may also be beneficial for sleep stage classification and seizure detection.

This study describes a method for classifying electrocorticograms (ECoGs) based on motor imagery (MI) on the brain&ndash, computer interface (BCI) system. This method is different from the traditional feature extraction and classification method. In this paper, the proposed method employs the deep learning algorithm for extracting features and the traditional algorithm for classification. Specifically, we mainly use the convolution neural network (CNN) to extract the features from the training data and then classify those features by combing with the gradient boosting (GB) algorithm. The comprehensive study with CNN and GB algorithms will profoundly help us to obtain more feature information from brain activities, enabling us to obtain the classification results from human body actions. The performance of the proposed framework has been evaluated on the dataset I of BCI Competition III. Furthermore, the combination of deep learning and traditional algorithms provides some ideas for future research with the BCI systems.

Brain&ndash, computer interfaces (BCIs) can help people with limited motor abilities to interact with their environment without external assistance. A major challenge in electroencephalogram (EEG)-based BCI development and research is the cross-subject classification of motor imagery data. Due to the highly individualized nature of EEG signals, it has been difficult to develop a cross-subject classification method that achieves sufficiently high accuracy when predicting the subject&rsquo, s intention. In this study, we propose a multi-branch 2D convolutional neural network (CNN) that utilizes different hyperparameter values for each branch and is more flexible to data from different subjects. Our model, EEGNet Fusion, achieves 84.1% and 83.8% accuracy when tested on the 103-subject eegmmidb dataset for executed and imagined motor actions, respectively. The model achieved statistically significantly higher results compared with three state-of-the-art CNN classifiers: EEGNet, ShallowConvNet, and DeepConvNet. However, the computational cost of the proposed model is up to four times higher than the model with the lowest computational cost used for comparison.

Left and right hand motor imagery electroencephalogram (MI-EEG) signals are widely used in brain-computer interface (BCI) systems to identify a participant intent in controlling external devices. However, due to a series of reasons, including low signal-to-noise ratios, there are great challenges for efficient motor imagery classification. The recognition of left and right hand MI-EEG signals is vital for the application of BCI systems. Recently, the method of deep learning has been successfully applied in pattern recognition and other fields. However, there are few effective deep learning algorithms applied to BCI systems, particularly for MI based BCI. In this paper, we propose an algorithm that combines continuous wavelet transform (CWT) and a simplified convolutional neural network (SCNN) to improve the recognition rate of MI-EEG signals. Using the CWT, the MI-EEG signals are mapped to time-frequency image signals. Then the image signals are input into the SCNN to extract the features and classify them. Tested by the BCI Competition IV Dataset 2b, the experimental results show that the average classification accuracy of the nine subjects is 83.2%, and the mean kappa value is 0.651, which is 11.9% higher than that of the champion in the BCI Competition IV. Compared with other algorithms, the proposed CWT-SCNN algorithm has a better classification performance and a shorter training time. Therefore, this algorithm could enhance the classification performance of MI based BCI and be applied in real-time BCI systems for use by disabled people.

Decoding motor imagery (MI) electroencephalogram (EEG) signals for brain-computer interfaces (BCIs) is a challenging task because of the severe non-stationarity of perceptual decision processes. Recently, deep learning techniques have had great success in EEG decoding because of their prominent ability to learn features from raw EEG signals automatically. However, the challenge that the deep learning method faces is that the shortage of labeled EEG signals and EEGs sampled from other subjects cannot be used directly to train a convolutional neural network (ConvNet) for a target subject. To solve this problem, in this paper, we present a novel conditional domain adaptation neural network (CDAN) framework for MI EEG signal decoding. Specifically, in the CDAN, a densely connected ConvNet is firstly applied to obtain high-level discriminative features from raw EEG time series. Then, a novel conditional domain discriminator is introduced to work as an adversarial with the label classifier to learn commonly shared intra-subjects EEG features. As a result, the CDAN model trained with sufficient EEG signals from other subjects can be used to classify the signals from the target subject efficiently. Competitive experimental results on a public EEG dataset (High Gamma Dataset) against the state-of-the-art methods demonstrate the efficacy of the proposed framework in recognizing MI EEG signals, indicating its effectiveness in automatic perceptual decision decoding.