Information Theoretic Approaches for Motor-Imagery BCI Systems: Review and Experimental Comparison

Surface Electromyography and Electroencephalogram-Based Gait Phase Recognition and Correlations Between Cortical and Locomotor Muscle in the Seven Gait Phases

The classification of gait phases based on surface electromyography (sEMG) and electroencephalogram (EEG) can be used to the control systems of lower limb exoskeletons for the rehabilitation of patients with lower limb disorders. In this study, the slope sign change (SSC) and mean power frequency (MPF) features of EEG and sEMG were used to recognize the seven gait phases [loading response (LR), mid-stance (MST), terminal stance (TST), pre-swing (PSW), initial swing (ISW), mid-swing (MSW), and terminal swing (TSW)]. Previous researchers have found that the cortex is involved in the regulation of treadmill walking. However, corticomuscular interaction analysis in a high level of gait phase granularity remains lacking in the time–frequency domain, and the feasibility of gait phase recognition based on EEG combined with sEMG is unknown. Therefore, the time–frequency cross mutual information (TFCMI) method was applied to research the theoretical basis of gait control in seven gait phases using beta-band EEG and sEMG data. We firstly found that the feature set comprising SSC of EEG as well as SSC and MPF of sEMG was robust for the recognition of seven gait phases under three different walking speeds. Secondly, the distribution of TFCMI values in eight topographies (eight muscles) was different at PSW and TSW phases. Thirdly, the differences of corticomuscular interaction between LR and MST and between TST and PSW of eight muscles were not significant. These insights enrich previous findings of the authors who have carried out gait phase recognition and provide a theoretical basis for gait recognition based on EEG and sEMG.

A neurophysiological approach to spatial filter selection for adaptive brain–computer interfaces

Objective. The common spatial patterns (CSP) algorithm is an effective method to extract discriminatory features from electroencephalography (EEG) to be used by a brain–computer interface (BCI). However, informed selection of CSP filters typically requires oversight from a BCI expert to accept or reject filters based on the neurophysiological plausibility of their activation patterns. Our goal was to identify, analyze and automatically classify prototypical CSP patterns to enhance the prediction of motor imagery states in a BCI. Approach. A data-driven approach that used four publicly available EEG datasets was adopted. Cluster analysis revealed recurring, visually similar CSP patterns and a convolutional neural network was developed to distinguish between established CSP pattern classes. Furthermore, adaptive spatial filtering schemes that utilize the categorization of CSP patterns were proposed and evaluated. Main results. Classes of common neurophysiologically probable and improbable CSP patterns were established. Analysis of the relationship between these categories of CSP patterns and classification performance revealed discarding neurophysiologically improbable filters can decrease decoder performance. Further analysis revealed that the spatial orientation of EEG modulations can evolve over time, and that the features extracted from the original CSP filters can become inseparable. Importantly, it was shown through a novel adaptive CSP technique that adaptation in response to these emerging patterns can restore feature separability. Significance. These findings highlight the importance of considering and reporting on spatial filter activation patterns in both online and offline studies. They also emphasize to researchers in the field the importance of spatial filter adaptation in BCI decoder design, particularly for online studies with a focus on training users to develop stable and suitable brain patterns.

Recognition of the idle state based on a novel IFB-OCN method for an asynchronous brain-computer interface

BACKGROUND

A major difficulty for the asynchronous brain-computer interface (BCI) lies in the accurate recognition of the control and idle states. Although subject's attention level was found to be different in these states, the validity of recognizing them using attention features has not been studied.

NEW METHODS

This paper proposed a novel Individualized Frequency Band based Optimized Complex Network (IFB-OCN) method to enhance the performance of discriminating the control and idle states. The IFB-OCN method extracted the attention features from a single FPz channel, selected the first three individualized frequency bands with the highest accuracies, and integrated the features of these bands for classification.

RESULTS

The performance was evaluated using a steady-state visual evoked potential (SSVEP)-based BCI task. In the offline evaluation, the IFB-OCN method achieved the highest average accuracy of 93.5 % with the data length of 4 s, and achieved the highest information transfer rate (ITR) of 47.3 bits/min with the data length of 0.5 s. In the simulated online evaluation, the IFB-OCN method obtained a true positive rate (TPR) of 89.8 % and a true negative rate (TNR) of 86.2 %.

COMPARISON WITH EXISTING METHODS

The proposed IFB-OCN method recognized the control and idle states using a single FPz channel rather than the occipital channels, and outperformed the existing algorithms in the accuracy of detecting the attention level.

CONCLUSIONS

These results demonstrate that the proposed IFB-OCN method is efficient in recognizing the idle state and has a great potential for enhancing the asynchronous BCIs.

Application of Continuous Wavelet Transform and Convolutional Neural Network in Decoding Motor Imagery Brain-Computer Interface

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.

Unsupervised Common Spatial Patterns

The common spatial pattern (CSP) method is a dimensionality reduction technique widely used in brain-computer interface (BCI) systems. In the two-class CSP problem, training data are linearly projected onto directions maximizing or minimizing the variance ratio between the two classes. The present contribution proves that kurtosis maximization performs CSP in an unsupervised manner, i.e., with no need for labeled data, when the classes follow Gaussian or elliptically symmetric distributions. Numerical analyses on synthetic and real data validate these findings in various experimental conditions, and demonstrate the interest of the proposed unsupervised approach.

EEG Signal Processing in MI-BCI Applications With Improved Covariance Matrix Estimators

In brain-computer interfaces (BCIs), the typical models of the EEG observations usually lead to a poor estimation of the trial covariance matrices, given the high non-stationarity of the EEG sources. We propose the application of two techniques that significantly improve the accuracy of these estimations and can be combined with a wide range of motor imagery BCI (MI-BCI) methods. The first one scales the observations in such a way that implicitly normalizes the common temporal strength of the source activities. When the scaling applies independently to the trials of the observations, the procedure justifies and improves the classical preprocessing for the EEG data. In addition, when the scaling is instantaneous and independent for each sample, the procedure particularizes to Tyler's method in statistics for obtaining a distribution-free estimate of scattering. In this case, the proposal provides an original interpretation of this existing method as a technique that pursuits an implicit instantaneous power-normalization of the underlying source processes. The second technique applies to the classifier and improves its performance through a convenient regularization of the features covariance matrix. Experimental tests reveal that a combination of the proposed techniques with the state-of-the-art algorithms for motor-imagery classification provides a significant improvement in the classification results.