Efficient Feature Learning Model of Motor Imagery EEG Signals with L1-Norm and Weighted Fusion

Abstract

Brain-computer interface (BCI) for motor imagery is an advanced technology used in the field of medical rehabilitation. However, due to the poor accuracy of electroencephalogram feature classification, BCI systems often misrecognize user commands. Although many state-of-the-art feature selection methods aim to enhance classification accuracy, they usually overlook the interrelationships between individual features, indirectly impacting the accuracy of feature classification. To overcome this issue, we propose an adaptive feature learning model that employs a Riemannian geometric approach to generate a feature matrix from electroencephalogram signals, serving as the model's input. By integrating the enhanced adaptive L1 penalty and weighted fusion penalty into the sparse learning model, we select the most informative features from the matrix. Specifically, we measure the importance of features using mutual information and introduce an adaptive weight construction strategy to penalize regression coefficients corresponding to each variable adaptively. Moreover, the weighted fusion penalty balances weight differences among correlated variables, reducing the model's overreliance on specific variables and enhancing accuracy. The performance of the proposed method was validated on BCI Competition IV datasets IIa and IIb using the support vector machine. Experimental results demonstrate the effectiveness and superiority of the proposed model compared to the existing models.

Keywords: electroencephalogram; feature selection; mutual information; penalty term; weighted fusion.

Figure 1

The overview of our proposed framework for motor imagery classification.

Figure 2

Timing paradigm of each trial. (a) BCI competition IV dataset IIa. (b) BCI competition IV dataset IIb.

Figure 3

Results of ablation experiments on dataset IIa.

Figure 4

Results of ablation experiments on dataset IIb.

Figure 5

The confusion matrix of the proposed model for subject 1 on dataset IIa.