Limb Motor Imagery Research Articles

The Paradigm Design of a Novel 2-class Unilateral Upper Limb Motor Imagery Tasks and its EEG Signal Classification.

Multitasking motor imagery (MI) of the unilateral upper limb is potentially more valuable in stroke rehabilitation than the current conventional MI in both hands. In this paper, a novel experimental paradigm was designed to imagine two motions of unilateral upper limb, which is hand gripping and releasing, and elbow reciprocating left and right. During this experiment, the electroencephalogram (EEG) signals were collected from 10 subjects. The time and frequency domains of the EEG signals were analyzed and visualized, indicating the presence of different Event-Related Desynchronization (ERD) or Event-Related Synchronization (ERS) for the two tasks. Then the two tasks were classified through three different EEG decoding methods, in which the optimized convolutional neural network (CNN) based on FBCNet achieved an average accuracy of 67.8%, obtaining a good recognition result. This work not only can advance the studies of MI decoding of unilateral upper limb, but also can provide a basis for better upper limb stroke rehabilitation in MI-BCI.

A linearly extendible multi-artifact removal approach for improved upper extremity EEG-based motor imagery decoding.

Non-invasive multichannel Electroencephalography (EEG) recordings provide an alternative source of neural information from which motor imagery (MI) patterns associated with limb movement intent can be decoded for use as control inputs for rehabilitation robots. The presence of multiple inherent dynamic artifacts in EEG signals, however, poses processing challenges for brain-computer interface (BCI) systems. A large proportion of the existing EEG signal preprocessing methods focus on isolating single artifact per time from an ensemble of EEG trials and require calibration and/or reference electrodes, resulting in increased complexity of their application to MI-EEG controlled rehabilitation devices in practical settings. Also, a few existing multi-artifacts removal methods though explored in other domains, they have rarely been investigated in the space of MI-EEG signals for multiple artifacts cancellation in a simultaneous manner. Building on the premise of previous works, this study propose a semi-automatic EEG preprocessing method that combines Generalized Eigenvalue Decomposition driven by low-rank approximation and a Multi-channel Wiener Filter (GEVD-MWF) that employs a learning technique for simultaneous elimination of multiple artifacts from MI-EEG signals. The proposed method is applied to remove multiple artifacts from 64-channel EEG signals recorded from transhumeral amputees while they performed distinct classes of upper limb MI tasks before decoding their movement intent using a selection of features and machine learning algorithms. Experimental results show that the proposed GEVD-MWF method yields significant improvements in MI decoding accuracies, in the range of 13.23%-41.21% compared to four existing popular artifact removal algorithms. Further investigation revealed that the GEVD-MWF approach enabled accuracies in the range of 90.44% - 99.67% using "single trial" EEG recordings, which could eliminate the need to record and process large ensembles of EEG trials as commonly required in some existing approaches. Additionally, using a variant of the sequential forward floating selection algorithm, a subset of 9 channels was used to obtain a decoding accuracy of 93.73%±1.58%. Given its improved performance, reduced data requirements, and feasibility with few channels, the proposed GEVD-MWF could potentially spur the development of effective real-time control strategies for multi-degree of freedom EEG-based miniaturized rehabilitation robotic interfaces.

A multi-modal modified feedback self-paced BCI to control the gait of an avatar

Brain–computer interfaces (BCIs) have been used to control the gait of a virtual self-avatar with a proposed application in the field of gait rehabilitation. Some limitations of existing systems are: (a) some systems use mental imagery (MI) of movements other than gait; (b) most systems allow the user to take single steps or to walk but do not allow both; (c) most function in a single BCI mode (cue-paced or self-paced). Objective. The objective of this study was to develop a high performance multi-modal BCI to control single steps and forward walking of an immersive virtual reality avatar. Approach. This system used MI of these actions, in cue-paced and self-paced modes. Twenty healthy participants participated in this study, which was comprised of four sessions across four different days. They were cued to imagine a single step forward with their right or left foot, or to imagine walking forward. They were instructed to reach a target by using the MI of multiple steps (self-paced switch-control mode) or by maintaining MI of forward walking (continuous-control mode). The movement of the avatar was controlled by two calibrated regularized linear discriminate analysis classifiers that used the µ power spectral density over the foot area of the motor cortex as a feature. The classifiers were retrained after every session. For a subset of the trials, positive modified feedback (MDF) was presented to half of the participants, where the avatar moved correctly regardless of the classification of the participants’ MI. The performance of the BCI was computed on each day, using different control modes. Main results. All participants were able to operate the BCI. Their average offline performance, after retraining the classifiers was 86.0 ± 6.1%, 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% showing that MDF enhanced BCI performance (p = 0.001). The average performance was 83% at self-paced switch control and 92% at continuous control mode. Significance. This study reports on a first BCI to use motor imagery of the lower limbs in order to control the gait of an avatar with different control modes and different control commands (single steps or forward walking). BCI performance is increased in a novel way by combining three different performance enhancement techniques, resulting in a single high performance and multi-modal BCI system. This study also showed that the improvements due to the effects of MDF lasted for more than one session.

EEG Feature Extraction Based on Brain Function Network and Sample Entropy

For the low recognition rate of motor imagery ElectroEncephaloGram (EEG) signals using single feature in Brain-Computer Interface (BCI) research, a feature extraction method combining brain function network and sample entropy is proposed. According to the neural mechanism appearing in Event Related Synchronization/Event Related Desynchronization (ERS/ERD) phenomenon and the contralateral mapping mechanism between cortex and limb motor imagery, the μ rhythm is denoised by wavelet packet transform. The brain function network is constructed for left hemispherical brain region and right hemispherical brain region by μ rhythm of 27 left channels and 27 right channels respectively. The mean node degree and the mean clustering coefficient are calculated as the brain function network characteristics, and the feature vectors combining the distribution and directivity are constructed by the sample entropy of C3 and C4 channels with the μ rhythm. The Support Vector Machine (SVM) is used to classify the left hand and right hand motor imagery EEG signals. The results show that the feature extraction method based on brain function network and sample entropy achieves better classification result, and the highest classification rate reached 90.27%.

A Novel Channel Selection Method for BCI Classification Using Dynamic Channel Relevance

Brain-Computer Interface (BCI) provides a direct communicating pathway between the human brain and the external environment. In the BCI systems, electroencephalography (EEG) signals are used to represent different cognitive patterns corresponding to various limb movements or motor imagery (MI) activities. However, EEG signals are multichannel in nature that require explicit information processing to alleviate the computational complexity of the BCI system. This paper represents a novel channel selection method to find effective EEG channels using Dynamic Channel Relevance (DCR) score. The proposed approach explores the minimum redundancy maximum relevance paradigm for selecting pertinent channels from raw ones. Firstly, the EEG signals are preprocessed using the Savitzky-Golay filter, and a sliding window approach is used to decompose them into chunks of fixed length. Further, three priorly known channels are used as candidate solutions for detecting new correlated channels. The selected channels are used to extract spatial-temporal features using the Multivariate Empirical Mode Decomposition (MEMD) method. The extracted features are used to discriminate four MI tasks (left hand, right hand, tongue, and foot) using the Support Vector Machine (SVM). The experiment is validated on three public EEG datasets (BCI Competition IV- 2008 – IIA, BCI Competition IV- dataset 1, BCI competition III - dataset IVa). The results show that the proposed method achieved a superior classification accuracy (85.4% on dataset 1, 80.33% on dataset 2, and 85.20% on dataset 3) with a lesser number of channels compared to state-of-the-art methods. In addition, our method significantly reduced the computation time compared to other published results without compromising the classification accuracy. Topographical mapping between the selected channels and cognitive regions showed that the central, frontal, and parietal lobe contributes to the execution of various MI tasks during physical activities.

Effect of power feature covariance shift on BCI spatial-filtering techniques: A comparative study

Background and ObjectiveThe input data distributions of EEG-based BCI systems can change during intra-session transitions due to nonstationarity caused by features covariate shifts, thus compromising BCI performance.We aimed to identify the most robust spatial filtering approach, among most used methods, testing them on calibration dataset, and test dataset recorded 30 min afterwards. In addition, we also investigated if their performance improved after application of Stationary Subspace Analysis (SSA). MethodsWe have recorded, in 17 healthy subjects, the calibration set at the beginning of the upper limb motor imagery BCI experiment and testing set separately 30 min afterwards. Both the calibration and test data were pre-processed and the BCI models were produced by using several spatial filtering approaches on the calibration set. Those models were subsequently evaluated on a test set. The differences between the accuracy estimated by cross-validation on the calibration dataset and the accuracy on the test dataset were investigated. The same procedure was performed with, and without SSA pre-processing step. ResultsA significant reduction in accuracy on the test dataset was observed for CSP, SPoC and SpecRCSP approaches. For SLap and SpecCSP only a slight decreasing trend was observed, while FBCSP and FBCSPT largely maintained moderately high median accuracy >70%. In the case of application of SSA pre-processing, the differences between accuracy observed on calibration and test dataset were reduced. In addition, accuracy values both on calibration and test set were slightly higher in case of SSA pre-processing and also in this case FBCSP and FBCSPT presented slightly better performance compared to other methods. ConclusionThe intrinsic signal nonstationarity characteristics, caused by covariance shifts of power features, reduced the accuracy of BCI model, therefore, suggesting that this evaluation framework should be considered for testing and simulating real life performance. FBCSP and FBSCPT approaches showed to be more robust to feature covariance shift. SSA can improve the models performance and reduce accuracy decline from calibration to test set.

Random matrix theory for analysing the brain functional network in lower limb motor imagery.

We use random matrix theory (RMT) to investigate the statistical properties of brain functional networks in lower limb motor imagery. Functional connectivity was calculated by Pearson correlation coefficient (PCC), mutual information (MTI) and phase locking value (PLV) extracted from EEG signals. We found that when the measured subjects imagined the movements of their lower limbs the spectral density as well as the level spacings displayed deviations from the random matrix prediction. In particular, a significant difference between the left and right foot imaginary movements was observed in the maximum eigenvalue from the PCC, which can provide a theoretical basis for further study on the classification of unilateral movement of lower limbs.

EEG-based Classification of Lower Limb Motor Imagery with Brain Network Analysis

This study aims to investigate the difference in cortical signal characteristics between the left and right foot imaginary movements and to improve the classification accuracy of the experimental tasks. Raw signals were gathered from 64-channel scalp electroencephalograms of 11 healthy participants. Firstly, the cortical source model was defined with 62 regions of interest over the sensorimotor cortex (nine Brodmann areas). Secondly, functional connectivity was calculated by phase lock value for α and β rhythm networks. Thirdly, network-based statistics were applied to identify whether there existed stable and significant subnetworks that formed between the two types of motor imagery tasks. Meanwhile, ten graph theory indices were investigated for each network by t-test to determine statistical significance between tasks. Finally, sparse multinomial logistic regression (SMLR)-support vector machine (SVM), as a feature selection and classification model, was used to analyze the graph theory features. The specific time–frequency (α event-related desynchronization and β event-related synchronization) difference network between the two tasks was congregated at the midline and demonstrated significant connections in the premotor areas and primary somatosensory cortex. A few of statistically significant differences in the network properties were observed between tasks in the α and β rhythm. The SMLR-SVM classification model achieved fair discrimination accuracy between imaginary movements of the two feet (maximum 75% accuracy rate in single-trial analyses). This study reveals the network mechanism of the discrimination of the left and right foot motor imagery, which can provide a novel avenue for the BCI system by unilateral lower limb motor imagery.

Age-related differences in postural adjustments during limb movement and motor imagery in young and older adults

Recent research has shown that systematic postural adjustments occur during periods of manual motor imagery (MI), but the timing (anticipatory or reactive) and directionality (against or in the direction of arm extension) of these postural motions relative to individual manual actions or imagery are not well understood. This study analyzed the anteroposterior hip and head motion of healthy young and older participants, while they imagined bilateral arm raises under self-initiated or environmentally triggered performance conditions. When MI was self-initiated, both age groups showed significant forward postural motion during the second prior to MI initiation. When MI (or physical arm movement) was environmentally triggered, however, older people did not show anticipatory forward postural motion, but did show compensatory backward head motion. These results suggest that manual MI is indeed accompanied by anticipatory postural motion, but this anticipation is attenuated in older people when they do not have control over the timing of manual movement onset.

Classifying electrical activity of the brain during imaginary movements of untrained subjects using artificial neural networks

Introduction: Developing new classification methods for human brain electrical activity patterns corresponding to actual movements or motor imagery is an essential interdisciplinary problem in brain-computer interface research. One of the most promising approaches is the development of methods based on artificial neural networks. Purpose: The development of ANN-based methods for classifying electroencephalographic patterns associated with motor imagery in untrained subjects. Methods: Classifiers based on linear neural networks, multi-layer perceptrons, radial basis function networks and support vector machines. Results: The authors selected the optimal type, topology, learning algorithms and parameters of an artificial neural network in order to provide the most accurate and fast classification of lower limb motor imagery EEG signals. It has been studied how the number of the analyzed channels of a multichannel EEG and their choice affect the quality of motor imagery patterns classification. Optimal configurations were obtained for the electrode arrangements. The influence of EEG pre-processing on the accuracy of motor imagery recognition was analyzed. A computational experiment showed the accuracy of 90-95% in untrained subjects. Radial basis function network demonstrated the best performance. Besides, the dataset dimensionality has been significantly reduced down to 6–12 channels without any classification accuracy loss. Practical relevance: The obtained results can be useful for the developers of motor imagery EEG classification algorithms used in brain-computer interfaces.

Cerebral hemodynamic changes during motor imagery and passive robot-assisted movement of the lower limbs

BackgroundNeurovascular Coupling is the cerebral mechanism responsible for linking neuronal activity, cerebral metabolism and regional cerebral blood flow (CBF). The direct relation between functional brain activity during active, passive and motor imagery paradigms and changes in CBF has been widely investigated using different techniques. However, CBF changes have not been investigated beat by beat during robot assisted passive movement (PM) and motor imagery (MI) of lower limb, yet. Materials and methodsWe investigated beat-to-beat hemodynamic changes in 8 healthy subjects using TCD during MI and robot-assisted PM of lower limb. ResultsThe results showed that MI and PM induce a significant CBFv increase and that PM and MI lead to similar hemodynamic changes in healthy subjects. ConclusionsThe findings may be useful to better understand the variation of CBFv in brain pathology and to develop more specific and efficient rehabilitation therapy protocols in neurological diseases, such as stroke.

Motor-Induced Suppression of the N100 Event-Related Potential During Motor Imagery Control of a Speech Synthesizer Brain-Computer Interface.

Purpose Speech motor control relies on neural processes for generating sensory expectations using an efference copy mechanism to maintain accurate productions. The N100 auditory event-related potential (ERP) has been identified as a possible neural marker of the efference copy with a reduced amplitude during active listening while speaking when compared to passive listening. This study investigates N100 suppression while controlling a motor imagery speech synthesizer brain-computer interface (BCI) with instantaneous auditory feedback to determine whether similar mechanisms are used for monitoring BCI-based speech output that may both support BCI learning through existing speech motor networks and be used as a clinical marker for the speech network integrity in individuals without severe speech and physical impairments. Method The motor-induced N100 suppression is examined based on data from 10 participants who controlled a BCI speech synthesizer using limb motor imagery. We considered listening to auditory target stimuli (without motor imagery) in the BCI study as passive listening and listening to BCI-controlled speech output (with motor imagery) as active listening since audio output depends on imagined movements. The resulting ERP was assessed for statistical significance using a mixed-effects general linear model. Results Statistically significant N100 ERP amplitude differences were observed between active and passive listening during the BCI task. Post hoc analyses confirm the N100 amplitude was suppressed during active listening. Conclusion Observation of the N100 suppression suggests motor planning brain networks are active as participants control the BCI synthesizer, which may aid speech BCI mastery.

DECODING OF SIMPLE AND COMPOUND LIMB MOTOR IMAGERY MOVEMENTS BY FRACTAL ANALYSIS OF ELECTROENCEPHALOGRAM (EEG) SIGNAL

One of the major attempts in rehabilitation science is to decode different movements of human using physiological signals. Since human movements are mainly controlled by the brain, decoding of movements by analysis of the brain activity has great importance. In this paper, we apply fractal analysis to Electroencephalogram (EEG) signal in order to decode simple and compound limb motor imagery movements. The fractal dimension of EEG signal is analyzed in case of left hand, right hand, both hands, feet, left hand combined with right foot, and right hand combined with left foot movements. Based on the obtained results, EEG signal experiences the lowest and greatest fractal dimension in case of both hands movement, and feet movement, respectively. Besides obtaining different fractal dimension for EEG signal in case of different movements, no significant difference was observed in fractal dimension of EEG signal between different movements. The method of analysis employed in this research can be widely applied to analysis of EEG signal for decoding of different movements of human.

Walking Imagery Evaluation in Brain Computer Interfaces via a Multi-View Multi-Level Deep Polynomial Network.

Brain-computer interfaces based on motor imagery (MI) have been widely used to support the rehabilitation of motor functions of the upper limbs rather than lower limbs. This is probably because it is more difficult to detect the brain activities of lower limb MI. In order to reliably detect the brain activities of lower limbs to restore or improve the walking ability of the disabled, we propose a new paradigm of walking imagery (WI) in a virtual environment (VE), in order to elicit the reliable brain activities and achieve a significant training effect. First, we extract and fuse both the spatial and time-frequency features as a multi-view feature to represent the patterns in the brain activity. Second, we design a multi-view multi-level deep polynomial network (MMDPN) to explore the complementarity among the features so as to improve the detection of walking from an idle state. Our extensive experimental results show that the VE-based paradigm significantly performs better than the traditional text-based paradigm. In addition, the VE-based paradigm can effectively help users to modulate the brain activities and improve the quality of electroencephalography signals. We also observe that the MMDPN outperforms other deep learning methods in terms of classification performance.

Riemann Kernel Support Vector Machine Recursive Feature Elimination in the Field of Compound Limb Motor Imagery BCI&#x0D;

Riemann Kernel Support Vector Machine Recursive Feature Elimination in the Field of Compound Limb Motor Imagery BCI&#x0D;

Improving Real-Time Lower Limb Motor Imagery Detection Using tDCS and an Exoskeleton.

The aim of this work was to test if a novel transcranial direct current stimulation (tDCS) montage boosts the accuracy of lower limb motor imagery (MI) detection by using a real-time brain-machine interface (BMI) based on electroencephalographic (EEG) signals. The tDCS montage designed was composed of two anodes and one cathode: one anode over the right cerebrocerebellum, the other over the motor cortex in Cz, and the cathode over FC2 (using the International 10–10 system). The BMI was designed to detect two MI states: relax and gait MI; and was based on finding the power at the frequency which attained the maximum power difference between the two mental states at each selected EEG electrode. Two different single-blind experiments were conducted, E1 and a pilot test E2. E1 was based on visual cues and feedback and E2 was based on auditory cues and a lower limb exoskeleton as feedback. Twelve subjects participated in E1, while four did so in E2. For both experiments, subjects were separated into two equally-sized groups: sham and active tDCS. The active tDCS group achieved 12.6 and 8.2% higher detection accuracy than the sham group in E1 and E2, respectively, reaching 65 and 81.6% mean detection accuracy in each experiment. The limited results suggest that the exoskeleton (E2) enhanced the detection of the MI tasks with respect to the visual feedback (E1), increasing the accuracy obtained in 16.7 and 21.2% for the active tDCS and sham groups, respectively. Thus, the small pilot study E2 indicates that using an exoskeleton in real-time has the potential of improving the rehabilitation process of cerebrovascular accident (CVA) patients, but larger studies are needed in order to further confirm this claim.

The Study of Visual-Auditory Interactions on Lower Limb Motor Imagery.

In order to improve the activation of the mirror neuron system and the ability of the visual-cued motor imagery further, the multi-stimuli-cued unilateral lower limb motor imagery is studied in this paper. The visual-auditory evoked pathway is proposed and the sensory process is studied. To analyze the visual-auditory interactions, the kinesthetic motor imagery with the visual-auditory stimulus, visual stimulus and no stimulus are involved. The motor-related rhythm suppression is applied on quantitative evaluation. To explore the statistical sensory process, the causal relationships among the functional areas and the event-related potentials are investigated. The results have demonstrated the outstanding performances of the visual-auditory evoked motor imagery on the improvement of the mirror neuron system activation and the motor imagery ability. Besides, the abundant information interactions among functional areas and the positive impacts of the auditory stimulus in the motor and the visual areas have been revealed. The possibility that the sensory processes evoked by the visual-auditory interactions differ from the one elicited by kinesthetic motor imagery, has also been indicated. This study will promisingly offer an efficient way to motor rehabilitation, thus favorable for hemiparesis and partial paralysis patients.

Comparison of different EEG signal analysis techniques for an offline lower limb motor imagery brain-computer interface.

The use of motion assistance devices improves the rehabilitation process of patients that have motor disabilities. In the case these devices are controlled by brain-machine interfaces, the rehabilitation process can be improved due to neuroplasticity. However, in the case of lower limb rehabilitation, the limited accuracy of the control algorithms is a serious difficulty to overcome. In this research, different EEG signal's processing techniques, based on motor imagery, are tested for a brain-computer interface in an offline scenario, in order to detect the limitations of the models previous to its realtime implementation. The results reveal that motor imagery is very dependent on the subject and that Stockwell Transform provides the best accuracy among the models tested.

Novel tDCS montage favors lower limb motor imagery detection.

This work studies a novel transcranial direct current stimulation (tDCS) montage to improve a brain-machine interface (BMI) lower limb motor imagery detection. The tDCS montage is composed by two anodes and one cathode. One anode is located over the motor cortex and the other one over the cerebellum. Ten healthy subjects participated in this experiment. They were randomly separated into two groups: sham, which received a fake stimulation, and active tDCS, which received a real stimulation. Each subject was experimented on five consecutive days. Results pointed out that there was a significant difference $(p < 0 .05)$ in the classification accuracy between the sham and the active tDCS group. On each of the five days of the experiment the active tDCS group achieved better accuracy results than the sham group: 4%, 10%, 10%, 9% and 7% higher respectively.

Discrimination of four class simple limb motor imagery movements for brain–computer interface

Abstract The discrimination of four simple limb motor imagery movements for brain-computer interface (BCI) applications is still challenging. This is because most of the movement imaginations have close spatial representations on the motor cortex area. Nevertheless, due to its potential applications in significant areas including BCI, solutions need to be formulated to overcome the task discrimination issues faced when a motor imagery movement approach is utilized. Feature extraction is one of the most important steps in any BCI system; as such, enhancement to the existing methods has been incorporated in this work. For this, we propose four-class movement imaginations of the right hand, left hand, right foot, and left foot, and develop feature extraction methods utilizing discrete wavelet transform (DWT) and empirical mode decomposition (EMD); in both methods, artificial neural network (ANN) was used as a classifier. Based on the processed electroencephalography (EEG) data recorded from eleven subjects, it can be seen that EMD features outperform DWT features; the average accuracy achieved by the EMD features is 90.02%, and 84.77% using the DWT features. EMD even performs better than DWT in discriminating the most challenging tasks involving the right foot and left foot imageries, whose EEG data were derived from the same Cz node of the motor cortex.