Subject-Specific Channel Selection Using Time Information for Motor Imagery Brain–Computer Interfaces

Abstract

Keeping a minimal number of channels is essential for designing a portable brain-computer interface system for daily usage. Most existing methods choose key channels based on spatial information without optimization of time segment for classification. This paper proposes a novel subject-specific channel selection method based on a criterion called F score to realize the parameterization of both time segment and channel positions. The F score is a novel simplified measure derived from Fisher's discriminant analysis for evaluating the discriminative power of a group of features. The experimental results on a standard dataset (BCI competition III dataset IVa) show that our method can efficiently reduce the number of channels (from 118 channels to 9 in average) without a decrease in mean classification accuracy. Compared to two state-ofthe-art methods in channel selection, our method leads to comparable or even better classification results with less selected channels.

Subject-specific time-frequency selection for multi-class motor imagery-based BCIs using few Laplacian EEG channels

“…Thus, FDA-type F-score relies on the Euclidean distance between class centers to evaluate the difference between classes, and employs the trace of the covariance matrix to estimate the variance within one class. FDA-type F-score, as a simplified measure, avoids estimating a projection direction in multi-dimensional FDA, and has been successfully used in twoclass BCI and motor recognition studies for channel and feature selection [21,22,33,42].…”

“…Although data driven spatial filtering algorithms, such as common spatial pattern (CSP) [18,19] and independent component analysis [20], can greatly improve the SNR of sensorimotor rhythms, they usually require a large number of EEG channels. Multi-channel EEG recording reduces the portability of daily use BCI and therefore constitutes a main drawback for end users [21,22]. Thus, many methods have been developed to reduce the number of channels in MI-based BCI by using machine learning techniques to select an optimal channel subset [21][22][23][24][25][26][27].…”

“…Multi-channel EEG recording reduces the portability of daily use BCI and therefore constitutes a main drawback for end users [21,22]. Thus, many methods have been developed to reduce the number of channels in MI-based BCI by using machine learning techniques to select an optimal channel subset [21][22][23][24][25][26][27]. Due to individual differences between subjects, the estimated optimal subsets of channels usually vary with subjects.…”

“…FDA-type F-score is a simplified measure based on Fisher discriminant analysis (FDA) for measuring the discriminative power of a group of features. In our previous studies, we have successfully applied this criterion to seek the optimal channel subset for two-class BCIs [21] and to select the best kinematic feature group for two-class motor recognition [33]. Based on our preliminary work for two-class cases, here we extended the application of FDA-type F-score to multi-class cases using a one-versus-rest strategy.…”

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

“…An extension to the multiclass problem can be found in [67]. Since the optimal frequency bands can vary from subject to subject, several alternative approaches have been proposed that combine the time-frequency characteristics of the EEG data [68,69] for improving the classification accuracy and reducing the number of electrodes [70]. …”