Study of MI-BCI classification method based on the Riemannian transform of personalized EEG spatiotemporal features

Abstract

<abstract> <p>Motor imagery (MI) is a traditional paradigm of brain-computer interface (BCI) and can assist users in creating direct connections between their brains and external equipment. The common spatial patterns algorithm is the most popular spatial filtering technique for collecting EEG signal features in MI-based BCI systems. Due to the defect that it only considers the spatial information of EEG signals and is susceptible to noise interference and other issues, its performance is diminished. In this study, we developed a Riemannian transform feature extraction method based on filter bank fusion with a combination of multiple time windows. First, we proposed the multi-time window data segmentation and recombination method by combining it with a filter group to create new data samples. This approach could capture individual differences due to the variation in time-frequency patterns across different participants, thereby improving the model's generalization performance. Second, Riemannian geometry was used for feature extraction from non-Euclidean structured EEG data. Then, considering the non-Gaussian distribution of EEG signals, the neighborhood component analysis (NCA) algorithm was chosen for feature selection. Finally, to meet real-time requirements and a low complexity, we employed a Support Vector Machine (SVM) as the classification algorithm. The proposed model achieved improved accuracy and robustness. In this study, we proposed an algorithm with superior performance on the BCI Competition IV dataset 2a, achieving an accuracy of 89%, a kappa value of 0.73 and an AUC of 0.9, demonstrating advanced capabilities. Furthermore, we analyzed data collected in our laboratory, and the proposed method achieved an accuracy of 77.4%, surpassing other comparative models. This method not only significantly improved the classification accuracy of motor imagery EEG signals but also bore significant implications for applications in the fields of brain-computer interfaces and neural engineering.</p> </abstract>

Personalized motor imagery prediction model based on individual difference of ERP

“…Classic MI [53] Sinc-EEGNet 0.73 Classic MI [53] EEGNet 0.64 Classic MI [49] FBCSP + SVM 0.76 Classic MI [49] AFBCSP + SVM 0.79 Classic MI (this study) CSP + DNN 0.62 AO-multi-DMPT (this study) CSP + DNN 0.79 MI action for the subject, thus avoiding a lengthy calibration period due to individual differences. The limitation is that the AO-multi-DMPT paradigm uses three groups of fixed actions, which prevents the subjects to achieve better MI-ACC by autonomous adjustments.…”

“…Many studies have shown that personalized approaches can effectively mitigate the negative effects of individual differences on MI-BCI performance. Based on the differences in the work, these studies are divided into two categories: the personalization paradigm [43][44][45][46] and the personalization algorithm [47][48][49][50].…”

“…In terms of the personalized algorithm, Wang et al achieved a personalized EEG feature extraction method based on filter banks and elastic networks [48]. Ding et al performed personalized classification via the Riemannian transform [49]. In addition, there are personalization methods like channel selection [51] and optimal parameters [52].…”