Search

In electroencephalography (EEG)-based affective brain–computer interfaces (aBCIs), there is a consensus that EEG features extracted from different frequency bands and channels have different abilities in emotion expression. Besides, EEG is so weak and non-stationary that easily causes distribution discrepancies for EEG data collected at different times; therefore, it is necessary to explore the affective activation patterns in cross-session emotion recognition. To address these two problems, we propose a self-weighted semi-supervised classification (SWSC) model in this article for joint EEG-based cross-session emotion recognition and affective activation patterns mining, whose merits include: 1) using both the labeled and unlabeled samples from different sessions for better capturing data characteristics; 2) introducing a self-weighted variable to learn the importance of EEG features adaptively and quantitatively; and 3) mining the activation patterns including the critical EEG frequency bands and channels automatically based on the learned self-weighted variable. Extensive experiments are conducted on the benchmark SEED_IV emotional dataset and SWSC obtained excellent average accuracies of 77.40%, 79.55%, and 81.52% in three cross-session emotion recognition tasks. Moreover, SWSC identifies that the Gamma frequency band contributes the most and the EEG channels in prefrontal, left/right temporal, and (central) parietal lobes are more important for cross-session emotion recognition. [ABSTRACT FROM AUTHOR]

• A tensor model of the dynamic brain functional network is proposed. • This method relies on the changing of the interaction of various brain regions. • Orthogonal decomposition of partially symmetric tensors can extract MI features. • The identification electrode is located near Cz during the MI period. The classification of motor imagery (MI) task based on Electroencephalography (EEG) is an important problem in brain-computer interface (BCI) system. The high-precision classification of MI is a challenging task in which the process of feature extraction is crucial step. In this work, we propose a tensor model of a dynamic brain functional network (DBFN) to decode motion intentions. First, we construct the brain functional network in each small window. Then, the BFN of each time window is superimposed into a DBFN tensor with time as the axis. A tensor decomposition method with orthogonal and partial symmetric constraints is used to analyze the DBFN. Finally, the core tensor features are used as an input of the extreme learning machine (ELM) for classification. The results show that the proposed method is better than the degree, clustering coefficient of network, and principal component analysis of DBFN matrix model and the average accuracies are improved by 17.33%, 12.91%, and 17.5% under ELM, respectively. Moreover, the classification accuracy of the proposed method has the lowest variance, i.e., 5.96, indicating that the core tensor features are more adaptable to the subjects. The proposed method has the highest accuracy of 95% under both ELM and support vector machine (SVM). The average accuracy rates of ELM and SVM are 87.08% and 85.83%, respectively. The proposed method effectively extracts the EEG signal characteristics of MI and has strong robustness. This provides a reference for further research on the feature extraction algorithm of BCI. [ABSTRACT FROM AUTHOR]

Background. Rapid serial visual presentation (RSVP) based brain-computer interface (BCI) is widely used to categorize the target and non-target images. The available information limits the prediction accuracy of single-trial using single-subject electroencephalography (EEG) signals. New Method. Hyperscanning is a new manner to record two or more subjects' signals simultaneously. So we designed a multi-level information fusion model for target image detection based on dual-subject RSVP, namely HyperscanNet. The two modules of this model fuse the data and features of the two subjects at the data and feature layers. A chunked long and short-term memory artificial neural network (LSTM) was used in the time dimension to extract features at different periods separately, completing fine-grained underlying feature extraction. While the feature layer is fused, some plain operations are used to complete the fusion of the data layer to ensure that important information is not missed. Results. Experimental results show that the F1-score (the harmonic mean of precision and recall) of this method with best group of channels and segment length is 82.76%. Comparison with existing methods. This method improves the F1-score by at least 5% compared to single-subject target detection. Conclusions. Target detection can be accomplished by the two subjects' collaboration to achieve a higher and more stable F1-score than a single subject. [Display omitted] • The ratio between target and non-target images is 1:9. • Designed a multi-level information fusion model for target image detection based on dual-subject RSVP. • Accomplished higher and more stable F1-score than single subject. [ABSTRACT FROM AUTHOR]