Temporal Combination Pattern Optimization Based on Feature Selection Method for Motor Imagery BCIs

Abstract

Common spatial pattern (CSP) method is widely used for spatial filtering and brain pattern extraction from electroencephalogram (EEG) signals in motor imagery (MI)-based brain-computer interfaces (BCIs). The participant-specific time window relative to the visual cue has a significant impact on the effectiveness of the CSP. However, the time window is usually selected experientially or manually. To solve this problem, we propose a novel feature selection approach for MI-based BCIs. Specifically, multiple time segments were obtained by decomposing each EEG sample of the MI task. Furthermore, the features were extracted by CSP from each time segment and were combined to form a new feature vector. Finally, the optimal temporal combination patterns for the new feature vector were selected based on four feature selection algorithms, i.e., mutual information, least absolute shrinkage and selection operator, principal component analysis and stepwise linear discriminant analysis (denoted as MUIN, LASSO, PCA, and SWLDA, respectively), and the classification algorithm was employed to evaluate the average classification accuracy. With three BCI competition datasets, the results of the four proposed algorithms were compared with traditional CSP algorithm in classification accuracy. Experimental results show that compared with traditional algorithm, the proposed methods significantly improve performance. Specifically, the LASSO achieved the highest accuracy (88.58%) among the proposed methods. Importantly, the average classification accuracies using the proposed approaches significantly improved 10.14% (MUIN), 11.40% (LASSO), 6.08% (PCA), and 10.25% (SWLDA) compared to that using CSP. These results indicate that the proposed approach is expected to be practical in MI-based BCIs.

Modulating Frustration and Agency Using Fabricated Input for Motor Imagery BCIs in Stroke Rehabilitation

“…Control of MI-based BCI systems have been improved and made more robust using various signal processing techniques for improving the signal-to-noise ratio, feature investigations,feature selection, and machine learning techniques see e.g. [20], [21], [29]- [37], but the control of an MI-based BCI could also be improved through proper training protocols adhering to universal learning principles, instructional design, and feedback [38], [39]. However, it may take time to learn to perform MI which may be abstract and new to many patients.…”

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

“…A selection of the temporal aspect (i.e. the starting points of the time windows of MI task for both training and testing samples) as proposed in [55,56] can further improve model accuracy and possibly make BCI models less affected by the power feature covariance shift.…”

A Tensor-Based Frequency Features Combination Method for Brain–Computer Interfaces

“…These p-values are corrected by FDR [42]. Moreover, we compablack our method with Jin's [13] and Jiang's [40] on dataset 2. As shown in Table . V, our results are comparable with Jiang's and much better than Jin's.…”