EEG sensorimotor rhythms' variation and functional connectivity measures during motor imagery: linear relations and classification approaches

A Transformer-Based Approach Combining Deep Learning Network and Spatial-Temporal Information for Raw EEG Classification

The attention mechanism of the Transformer has the advantage of extracting feature correlation in the long-sequence data and visualizing the model. As time-series data, the spatial and temporal dependencies of the EEG signals between the time points and the different channels contain important information for accurate classification. So far, Transformer-based approaches have not been widely explored in motor-imagery EEG classification and visualization, especially lacking general models based on cross-individual validation. Taking advantage of the Transformer model and the spatial-temporal characteristics of the EEG signals, we designed Transformer-based models for classifications of motor imagery EEG based on the PhysioNet dataset. With 3s EEG data, our models obtained the best classification accuracy of 83.31%, 74.44%, and 64.22% on two-, three-, and four-class motor-imagery tasks in cross-individual validation, which outperformed other state-of-the-art models by 0.88%, 2.11%, and 1.06%. The inclusion of the positional embedding modules in the Transformer could improve the EEG classification performance. Furthermore, the visualization results of attention weights provided insights into the working mechanism of the Transformer-based networks during motor imagery tasks. The topography of the attention weights revealed a pattern of event-related desynchronization (ERD) which was consistent with the results from the spectral analysis of Mu and beta rhythm over the sensorimotor areas. Together, our deep learning methods not only provide novel and powerful tools for classifying and understanding EEG data but also have broad applications for brain-computer interface (BCI) systems.

Subject-Dependent Artifact Removal for Enhancing Motor Imagery Classifier Performance under Poor Skills

The Electroencephalography (EEG)-based motor imagery (MI) paradigm is one of the most studied technologies for Brain-Computer Interface (BCI) development. Still, the low Signal-to-Noise Ratio (SNR) poses a challenge when constructing EEG-based BCI systems. Moreover, the non-stationary and nonlinear signal issues, the low-spatial data resolution, and the inter- and intra-subject variability hamper the extraction of discriminant features. Indeed, subjects with poor motor skills have difficulties in practicing MI tasks against low SNR scenarios. Here, we propose a subject-dependent preprocessing approach that includes the well-known Surface Laplacian Filtering and Independent Component Analysis algorithms to remove signal artifacts based on the MI performance. In addition, power- and phase-based functional connectivity measures are studied to extract relevant and interpretable patterns and identify subjects of inefficency. As a result, our proposal, Subject-dependent Artifact Removal (SD-AR), improves the MI classification performance in subjects with poor motor skills. Consequently, electrooculography and volume-conduction EEG artifacts are mitigated within a functional connectivity feature-extraction strategy, which favors the classification performance of a straightforward linear classifier.

Functional Connectivity Analysis in Motor-Imagery Brain Computer Interfaces

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.

Effective Connectivity for Decoding Electroencephalographic Motor Imagery Using a Probabilistic Neural Network

Motor imagery (MI)-based brain-computer interfaces have gained much attention in the last few years. They provide the ability to control external devices, such as prosthetic arms and wheelchairs, by using brain activities. Several researchers have reported the inter-communication of multiple brain regions during motor tasks, thus making it difficult to isolate one or two brain regions in which motor activities take place. Therefore, a deeper understanding of the brain's neural patterns is important for BCI in order to provide more useful and insightful features. Thus, brain connectivity provides a promising approach to solving the stated shortcomings by considering inter-channel/region relationships during motor imagination. This study used effective connectivity in the brain in terms of the partial directed coherence (PDC) and directed transfer function (DTF) as intensively unconventional feature sets for motor imagery (MI) classification. MANOVA-based analysis was performed to identify statistically significant connectivity pairs. Furthermore, the study sought to predict MI patterns by using four classification algorithms-an SVM, KNN, decision tree, and probabilistic neural network. The study provides a comparative analysis of all of the classification methods using two-class MI data extracted from the PhysioNet EEG database. The proposed techniques based on a probabilistic neural network (PNN) as a classifier and PDC as a feature set outperformed the other classification and feature extraction techniques with a superior classification accuracy and a lower error rate. The research findings indicate that when the PDC was used as a feature set, the PNN attained the greatest overall average accuracy of 98.65%, whereas the same classifier was used to attain the greatest accuracy of 82.81% with the DTF. This study validates the activation of multiple brain regions during a motor task by achieving better classification outcomes through brain connectivity as compared to conventional features. Since the PDC outperformed the DTF as a feature set with its superior classification accuracy and low error rate, it has great potential for application in MI-based brain-computer interfaces.

Neural component analysis: A spatial filter for electroencephalogram analysis

BACKGROUND

Spatial filtering and source separation are valuable tools in the analysis of EEG data. However, despite the well-known spatial localisation of individual cognitive processes within the brain, the available methods for source separation, such as the widely used blind source separation technique, do not take into account the spatial distributions and locations of sources. This can result in sub-optimal source identification.

NEW METHOD

We present a new method for deriving a spatial filter for EEG data that attempts to identify sources that are maximally spatially distinct from one another in terms of the spatial distributions of their projections.

RESULTS

We first evaluate our method with simulated EEG and show that it is able to separate EEG signals into components with distinct spatial distributions that closely resemble the original simulated sources. We also evaluate our method with real EEG and show it is able to identify a spatial filter that can be used to significantly improve classification accuracy of the P300 event-related potential (ERP).

COMPARISON WITH EXISTING METHODS

We compare our method to a state of the art blind source separation methods, fast independent component analysis (ICA) and common spatial patterns (CSP). We evaluate the methods suitability for a common source separation application, analysis of ERPs.

CONCLUSIONS

Our results show that our method is well suited to identifying spatial filters for EEG analysis. This has potential applications in a wide range of EEG signal processing applications.

Task-independent Electrophysiological Correlates of Motor Imagery Ability from Kinaesthetic and Visual Perspectives

Motor imagery (MI) ability is highly subjective, as indicated by the individual scores of the MIQ-3 questionnaire, and poor imagers compensate for the difficulty in performing MI with larger cerebral activations, as demonstrated by MI studies involving hands/limbs. In order to identify general, task-independent MI ability correlates, 16 volunteers were stratified with MIQ-3. The scores in the kinaesthetic (K) and 1st-person visual (V) perspectives were associated with EEG patterns obtained during K-MI and V-MI of the same complex MIQ-3 movements during these MI tasks (Spearman's correlation, significance at <0.05, SnPM corrected). EEG measures were relative to rest (relaxation, closed eyes), and based on six electrode clusters both for band spectral content and connectivity (Granger causality). Lower K-MI ability was associated with greater theta decreases during tasks in fronto-central clusters and greater inward information flow to prefrontal clusters for theta, high alpha and beta bands. On the other hand, power band relative decreases were associated with V-MI ability in fronto-central clusters for low alpha and left fronto-central and both centro-parietal clusters for beta bands. The results thus suggest different computational mechanisms for MI-V and MI-K. The association between low alpha/beta desynchronization and V-MIQ scores and between theta changes and K-MIQ scores suggest a cognitive effort with greater cerebral activation in participants with lower V-MI ability. The association between information flow to prefrontal hub and K-MI ability suggest the need for a continuous update of information to support MI-related executive functions in subjects with poor K-MI ability.

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.

Differentiating resting brain states using ordinal symbolic analysis

Symbolic methods of analysis are valuable tools for investigating complex time-dependent signals. In particular, the ordinal method defines sequences of symbols according to the ordering in which values appear in a time series. This method has been shown to yield useful information, even when applied to signals with large noise contamination. Here, we use ordinal analysis to investigate the transition between eyes closed (EC) and eyes open (EO) resting states. We analyze two electroencephalography datasets (with 71 and 109 healthy subjects) with different recording conditions (sampling rates and the number of electrodes in the scalp). Using as diagnostic tools the permutation entropy, the entropy computed from symbolic transition probabilities, and an asymmetry coefficient (that measures the asymmetry of the likelihood of the transitions between symbols), we show that the ordinal analysis applied to the raw data distinguishes the two brain states. In both datasets, we find that, during the EC-EO transition, the EO state is characterized by higher entropies and lower asymmetry coefficient, as compared to the EC state. Our results thus show that these diagnostic tools have the potential for detecting and characterizing changes in time-evolving brain states.

Motor imagery training induces changes in brain neural networks in stroke patients

Motor imagery is the mental representation of an action without overt movement or muscle activation. However, the effects of motor imagery on stroke-induced hand dysfunction and brain neural networks are still unknown. We conducted a randomized controlled trial in the China Rehabilitation Research Center. Twenty stroke patients, including 13 males and 7 females, 32–51 years old, were recruited and randomly assigned to the traditional rehabilitation treatment group (PP group, n = 10) or the motor imagery training combined with traditional rehabilitation treatment group (MP group, n = 10). All patients received rehabilitation training once a day, 45 minutes per session, five times per week, for 4 consecutive weeks. In the MP group, motor imagery training was performed for 45 minutes after traditional rehabilitation training, daily. Action Research Arm Test and the Fugl-Meyer Assessment of the upper extremity were used to evaluate hand functions before and after treatment. Transcranial magnetic stimulation was used to analyze motor evoked potentials in the affected extremity. Diffusion tensor imaging was used to assess changes in brain neural networks. Compared with the PP group, the MP group showed better recovery of hand function, higher amplitude of the motor evoked potential in the abductor pollicis brevis, greater fractional anisotropy of the right dorsal pathway, and an increase in the fractional anisotropy of the bilateral dorsal pathway. Our findings indicate that 4 weeks of motor imagery training combined with traditional rehabilitation treatment improves hand function in stroke patients by enhancing the dorsal pathway. This trial has been registered with the Chinese Clinical Trial Registry (registration number: ChiCTR-OCH-12002238).