White Matter Connectivity Pattern Associate with Characteristics of Scalp EEG Signals.

The rhythm of electroencephalogram (EEG) depends on the neuroanatomical-based parameters such as white matter (WM) connectivity. However, the impacts of these parameters on the specific characteristics of EEG have not been clearly understood. Previous studies demonstrated that, these parameters contribute the inter-subject differences of EEG during performance of specific task such as motor imagery (MI). Though researchers have worked on this phenomenon, the idea is yet to be understood in terms of the mechanism that underlies such differences. Here, to tackle this issue, we began our investigations by first examining the structural features related to scalp EEG characteristics, which are event-related desynchronizations (ERDs), during MI using diffusion MRI. Twenty-four right-handed subjects were recruited to accomplish MI tasks and MRI scans. Based on the high spatial resolution of the structural and diffusion images, the motor-related WM links, such as basal ganglia (BG)-primary somatosensory cortex (SM1) pathway and supplementary motor area (SMA)-SM1 connection, were reconstructed by using probabilistic white matter tractography. Subsequently, the relationships of WM characteristics with EEG signals were investigated. These analyses demonstrated that WM pathway characteristics, including the connectivity strength and the positional characteristics of WM connectivity on SM1 (defined by the gyrus-sulcus ratio of connectivity, GSR), have a significant impact on ERDs when doing MI. Interestingly, the high GSR of WM connections between SM1 and BG were linked to the better ERDs. These results therefore, indicated that the connectivity in the gyrus of SM1 interacted with MI network which played the critical role for the scalp EEG signal extraction of MI to a great extent. The study provided the coupling mechanism between structural and dynamic physiological features of human brain, which would also contribute to understanding individual differences of EEG in MI-brain computer interface.

Predicting Inter-session Performance of SMR-Based Brain-Computer Interface Using the Spectral Entropy of Resting-State EEG.

Currently most subjects can control the sensorimotor rhythm-based brain-computer interface (SMR-BCI) successfully after several training procedures. However, 15-30% of subjects cannot achieve SMR-BCI control even after long-term training, and they are termed as "BCI inefficiency". This study focuses on the investigation of reliable SMR-BCI performance predictor. 40 subjects participated in the first experimental session and 26 of them returned in the second session, each session consists of an eyes closed/open resting-state EEG recording run and four EEG recording runs with hand motor imagery. We found spectral entropy derived from eyes closed resting-state EEG of channel C3 has a high correlation with SMR-BCI performance (r = 0.65). Thus, we proposed to use it as a biomarker to predict individual SMR-BCI performance. Receiver operating characteristics analysis and leave-one-out cross-validation demonstrated that the spectral entropy predictor provide outstanding classification capability for high and low aptitude BCI users. To our knowledge, there has been no discussion about the reliability of inter-session prediction in previous studies. We further evaluated the inter-session prediction performance of the spectral entropy predictor, and the results showed that the average classification accuracy of inter-session prediction up to 89%. The proposed predictor is convenient to obtain because it derived from single channel resting-state EEG, it could be used to identify potential SMR-BCI inefficiency subjects from novel users. But there are still limitations because Kübler et al. have shown that some BCI users may need eight or more sessions before they develop classifiable SMR activity.

L1 norm based common spatial patterns decomposition for scalp EEG BCI.

BACKGROUND: Brain computer interfaces (BCI) is one of the most popular branches in biomedical engineering. It aims at constructing a communication between the disabled persons and the auxiliary equipments in order to improve the patients' life. In motor imagery (MI) based BCI, one of the popular feature extraction strategies is Common Spatial Patterns (CSP). In practical BCI situation, scalp EEG inevitably has the outlier and artifacts introduced by ocular, head motion or the loose contact of electrodes in scalp EEG recordings. Because outlier and artifacts are usually observed with large amplitude, when CSP is solved in view of L2 norm, the effect of outlier and artifacts will be exaggerated due to the imposing of square to outliers, which will finally influence the MI based BCI performance. While L1 norm will lower the outlier effects as proved in other application fields like EEG inverse problem, face recognition, etc. METHODS: In this paper, we present a new CSP implementation using the L1 norm technique, instead of the L2 norm, to solve the eigen problem for spatial filter estimation with aim to improve the robustness of CSP to outliers. To evaluate the performance of our method, we applied our method as well as the standard CSP and the regularized CSP with Tikhonov regularization (TR-CSP), on both the peer BCI dataset with simulated outliers and the dataset from the MI BCI system developed in our group. The McNemar test is used to investigate whether the difference among the three CSPs is of statistical significance. RESULTS: The results of both the simulation and real BCI datasets consistently reveal that the proposed method has much higher classification accuracies than the conventional CSP and the TR-CSP. CONCLUSIONS: By combining L1 norm based Eigen decomposition into Common Spatial Patterns, the proposed approach can effectively improve the robustness of BCI system to EEG outliers and thus be potential for the actual MI BCI application, where outliers are inevitably introduced into EEG recordings.

Local temporal correlation common spatial patterns for single trial EEG classification during motor imagery.

Common spatial pattern (CSP) is one of the most popular and effective feature extraction methods for motor imagery-based brain-computer interface (BCI), but the inherent drawback of CSP is that the estimation of the covariance matrices is sensitive to noise. In this work, local temporal correlation (LTC) information was introduced to further improve the covariance matrices estimation (LTCCSP). Compared to the Euclidean distance used in a previous CSP variant named local temporal CSP (LTCSP), the correlation may be a more reasonable metric to measure the similarity of activated spatial patterns existing in motor imagery period. Numerical comparisons among CSP, LTCSP, and LTCCSP were quantitatively conducted on the simulated datasets by adding outliers to Dataset IVa of BCI Competition III and Dataset IIa of BCI Competition IV, respectively. Results showed that LTCCSP achieves the highest average classification accuracies in all the outliers occurrence frequencies. The application of the three methods to the EEG dataset recorded in our laboratory also demonstrated that LTCCSP achieves the highest average accuracy. The above results consistently indicate that LTCCSP would be a promising method for practical motor imagery BCI application.

Z-score linear discriminant analysis for EEG based brain-computer interfaces.

Linear discriminant analysis (LDA) is one of the most popular classification algorithms for brain-computer interfaces (BCI). LDA assumes Gaussian distribution of the data, with equal covariance matrices for the concerned classes, however, the assumption is not usually held in actual BCI applications, where the heteroscedastic class distributions are usually observed. This paper proposes an enhanced version of LDA, namely z-score linear discriminant analysis (Z-LDA), which introduces a new decision boundary definition strategy to handle with the heteroscedastic class distributions. Z-LDA defines decision boundary through z-score utilizing both mean and standard deviation information of the projected data, which can adaptively adjust the decision boundary to fit for heteroscedastic distribution situation. Results derived from both simulation dataset and two actual BCI datasets consistently show that Z-LDA achieves significantly higher average classification accuracies than conventional LDA, indicating the superiority of the new proposed decision boundary definition strategy.