Regularized common spatial patterns with subject-to-subject transfer of EEG signals

Abstract

In the context of brain-computer interface (BCI) system, the common spatial patterns (CSP) method has been used to extract discriminative spatial filters for the classification of electroencephalogram (EEG) signals. However, the classification performance of CSP typically deteriorates when a few training samples are collected from a new BCI user. In this paper, we propose an approach that maintains or improves the recognition accuracy of the system with only a small size of training data set. The proposed approach is formulated by regularizing the classical CSP technique with the strategy of transfer learning. Specifically, we incorporate into the CSP analysis inter-subject information involving the same task, by minimizing the difference between the inter-subject features. Experimental results on two data sets from BCI competitions show that the proposed approach greatly improves the classification performance over that of the conventional CSP method; the transformed variant proved to be successful in almost every case, based on a small number of available training samples.

Keywords: Brain-computer interfaces (BCI); Common spatial pattern (CSP); Electroencephalogram (EEG); Motor imagery (MI); Transfer learning.

Fig. 1

Average classification rates (%), as well as standard deviations, with varying numbers of training trials of the target subject for the five subjects on data set IVa of BCI competition III. a subject A1, b subject A2, c subject A3, d subject A4, e subject A5. In each panel, the first column denotes the case that none training trials of the target subject were used while the last column is the accuracy in the case of using all the training trial of the target subject in the training process. Each point represents the mean (±SD) of multiple determinations