Distribution Adaptation and Classification Framework Based on Multiple Kernel Learning for Motor Imagery BCI Illiteracy

Abstract

A brain-computer interface (BCI) translates a user’s thoughts such as motor imagery (MI) into the control of external devices. However, some people, who are defined as BCI illiteracy, cannot control BCI effectively. The main characteristics of BCI illiterate subjects are low classification rates and poor repeatability. To address the problem of MI-BCI illiteracy, we propose a distribution adaptation method based on multi-kernel learning to make the distribution of features between the source domain and target domain become even closer to each other, while the divisibility of categories is maximized. Inspired by the kernel trick, we adopted a multiple-kernel-based extreme learning machine to train the labeled source-domain data to find a new high-dimensional subspace that maximizes data divisibility, and then use multiple-kernel-based maximum mean discrepancy to conduct distribution adaptation to eliminate the difference in feature distribution between domains in the new subspace. In light of the high dimension of features of MI-BCI illiteracy, random forest, which can effectively handle high-dimensional features without additional cross-validation, was employed as a classifier. The proposed method was validated on an open dataset. The experimental results show that that the method we proposed suits MI-BCI illiteracy and can reduce the inter-domain differences, resulting in a reduction in the performance degradation of both cross-subjects and cross-sessions.

Paving the Way for Motor Imagery-Based Tele-Rehabilitation through a Fully Wearable BCI System

“…A limitation of this study in tele-rehabilitation applications is that the multimodal proposed feedback was positively biased. Nonetheless, this can be enhanced with an adaptive bias to optimize system performance and patient learning [62], and future development could also focus on improving the classification algorithm to enhance performance across sessions and deliver better feedback [63]. Although multiple sessions were carried out already with healthy subjects, it is worth emphasizing patients would require even more training sessions to gain proper control over the BCI system and obtain benefits from therapy [64].…”

Bridging the BCI illiteracy gap: a subject-to-subject semantic style transfer for EEG-based motor imagery classification

“…In order to tackle the BCI illiteracy problem, various machine learning (ML)-based approaches (Vidaurre and Blankertz, 2010 ; Vidaurre et al, 2011 ; Tao et al, 2022 ) have been developed. One such approach is co-adaptive learning (Vidaurre and Blankertz, 2010 ), which uses the ML algorithm, i.e., linear discriminant analysis (LDA), to help users achieve closed-loop feedback.…”

“…During the feedback process, both the user and the ML algorithm adapt to each other, thereby improving the overall performance of the BCI system. Another ML-based approach is based on multi-kernel learning (Tao et al, 2022 ) that aims to make the distribution of features closer to each other, while maximizing the divisibility of categories. Despite the reasonable performance achieved by ML-based approaches, they often rely on heuristic statistical reasoning and assumptions such as linear separability (Medin and Schwanenflugel, 1981 ) and same feature space assumption (Girolami, 2002 ).…”