The common spatial pattern (CSP) algorithm is a well-recognized spatial filtering method for feature extraction in motor imagery (MI)-based brain–computer interfaces (BCIs). However, due to the influence of nonstationary in electroencephalography (EEG) and inherent defects of the CSP objective function, the spatial filters, and their corresponding features are not necessarily optimal in the feature space used within CSP. In this work, we design a new feature selection method to address this issue by selecting features based on an improved objective function. Especially, improvements are made in suppressing outliers and discovering features with larger interclass distances. Moreover, a fusion algorithm based on the Dempster–Shafer theory is proposed, which takes into consideration the distribution of features. With two competition data sets, we first evaluate the performance of the improved objective functions in terms of classification accuracy, feature distribution, and embeddability. Then, a comparison with other feature selection methods is carried out in both accuracy and computational time. Experimental results show that the proposed methods consume less additional computational cost and result in a significant increase in the performance of MI-based BCI systems.

A potential limitation of a motor imagery (MI) based brain-computer interface (BCI) is that it usually requires a relatively long time to record sufficient EEG data for robust classifier training. The calibration burden during data acquisition phase will most probably cause a subject to be reluctant to use a BCI system. To alleviate this issue, we propose a novel sparse group representation model (SGRM) for improving the efficiency of MI-based BCI by exploiting the inter-subject information. Specifically, preceded by feature extraction using common spatial pattern, a composite dictionary matrix is constructed with training samples from both the target subject and other subjects. By explicitly exploiting within-group sparse and group-wise sparse constraints, the most compact representation of a test sample of the target subject is then estimated as a linear combination of columns in the dictionary matrix. Classification is implemented by calculating the class-specific representation residual based on the significant training samples corresponding to the nonzero representation coefficients. Accordingly, the proposed SGRM method effectively reduces the required training samples from the target subject due to auxiliary data available from other subjects. With two public EEG datasets, extensive experimental comparisons are carried out between SGRM and other state-of-the-art approaches. Superior classification performance of our method using 40 trials of the target subject for model calibration (Averaged accuracy = 78.2%, Kappa = 0.57 and Averaged accuracy = 77.7%, Kappa = 0.55 for the two datasets, respectively) indicates its promising potential for improving the practicality of MI-based BCI.

One of the most common types of brain-computer interfaces (BCIs) is called a P300 BCI, since it relies on the P300 and other event-related potentials (ERPs). In the canonical P300 BCI approach, items on a monitor flash briefly to elicit the necessary ERPs. Very recent work has shown that this approach may yield lower performance than alternate paradigms in which the items do not flash but instead change in other ways, such as moving, changing colour or changing to characters overlaid with faces. The present study sought to extend this research direction by parametrically comparing different ways to change items in a P300 BCI. Healthy subjects used a P300 BCI across six different conditions. Three conditions were similar to our prior work, providing the first direct comparison of characters flashing, moving, and changing to faces. Three new conditions also explored facial motion and emotional expression. The six conditions were compared across objective measures such as classification accuracy and bit rate as well as subjective measures such as perceived difficulty. In line with recent studies, our results indicated that the character flash condition resulted in the lowest accuracy and bit rate. All four face conditions (mean accuracy >91%) yielded significantly better performance than the flash condition (mean accuracy = 75%). Objective results reaffirmed that the face paradigm is superior to the canonical flash approach that has dominated P300 BCIs for over 20 years. The subjective reports indicated that the conditions that yielded better performance were not considered especially burdensome. Therefore, although further work is needed to identify which face paradigm is best, it is clear that the canonical flash approach should be replaced with a face paradigm when aiming at increasing bit rate. However, the face paradigm has to be further explored with practical applications particularly with locked-in patients.

The performance of motor imagery (MI) based Brain-computer interfacing (BCI) is easily affected by noise and redundant information that exists in the multi-channel electroencephalogram (EEG). To solve this problem, many temporal and spatial feature based channel selection methods have been proposed. However, temporal and spatial features do not accurately reflect changes in the power of the oscillatory EEG. Thus, spectral features of MI-related EEG signals may be useful for channel selection. Bispectrum analysis is a technique developed for extracting non-linear and non-Gaussian information from non-linear and non-Gaussian signals. The features extracted from bispectrum analysis can provide frequency domain information about the EEG. Therefore, in this study, we propose a bispectrumbased channel selection (BCS) method for MI-based BCI. The proposed method uses the sum of logarithmic amplitudes (SLA) and the first order spectral moment (FOSM) features extracted from bispectrum analysis to select EEG channels without redundant information. Three public BCI competition datasets (BCI competition IV dataset 1, BCI competition III dataset IVa and BCI competition III dataset IIIa) were used to validate the effectiveness of our proposed method. The results indicate that our BCS method outperforms use of all channels (83.8% vs 69.4%, 86.3% vs 82.9% and 77.8% vs 68.2%, respectively). Furthermore, compared to the other state-of-the-art methods, our BCS method also can achieve significantly better classification accuracies for MI-based BCI (Wilcoxon signed test, p<0.05).