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]

Extreme learning machine (ELM) is an emerging learning algorithm for the generalized single hidden layer feedforward neural networks in which the parameters of hidden units are randomly generated and thus the output weights can be analytically calculated. From the hidden to output layer, ELM essentially learns the output weight matrix based on the least squares regression formula that can be used for both classification/regression and dimensionality reduction. In this paper, we impose the orthogonal constraint on the output weight matrix and then formulate an orthogonal extreme learning machine (OELM) model, which produces orthogonal basis functions and can have more locality preserving power from ELM feature space to output layer than ELM. Since the locality preserving ability is potentially related to the discriminating power, the OELM is expect to have more discriminating power than ELM. Considering the case that the number of hidden units is usually greater than the number of classes, we propose an effective method to optimize the OELM objective by solving an orthogonal procrustes problem. Experiments by pairwisely comparing OELM with ELM on three widely used image data sets show the effectiveness of learning orthogonal mapping especially when given only limited training samples. [ABSTRACT FROM AUTHOR]

Driver fatigue has been under more attention as it is a main cause of traffic accidents. This paper proposed a method which utilized the inter/intra-region phase synchronization and functional units (FUs) to explore whether EEG synchronization changes from the alert state to the fatigue state. Mean phase coherence (MPC) is adopted as a measure for the phase synchronization. In order to find spatial-frequency features associated with mental state, we studied the intra/inter-region phase synchronization of EEG in different frequencies. The major finding is that EEG synchronizations in delta and alpha bands in frontal and parietal lobe are significantly increased as the mental state of the driver shifted from alertness to fatigue. This finding is simultaneously validated by NASA-Task Load Index (TLX) and Karolinska sleepiness scale (KSS). The statistical analysis results suggest MPC may be used to distinguish between alert and fatigue state of mind. In addition, the another contribution of the work indicates a simple and significant spatial-frequency pair of electrodes, i.e., Fz-Oz in delta band, to evaluate driver fatigue. It helps to implement real-world applications with wearable EEG equipment. [ABSTRACT FROM AUTHOR]

Various relations existing in Electroencephalogram (EEG) data are significant for EEG feature representation. Thus, studies on the graph-based method focus on extracting relevancy between EEG channels. The shortcoming of existing graph studies is that they only consider a single relationship of EEG electrodes, which results an incomprehensive representation of EEG data and relatively low accuracy of emotion recognition. In this paper, we propose a fusion graph convolutional network (FGCN) to extract various relations existing in EEG data and fuse these extracted relations to represent EEG data more comprehensively for emotion recognition. First, the FGCN mines brain connection features on topology, causality, and function. Then, we propose a local fusion strategy to fuse these three graphs to fully utilize the valuable channels with strong topological, causal, and functional relations. Finally, the graph convolutional neural network is adopted to represent EEG data for emotion recognition better. Experiments on SEED and SEED-IV demonstrate that fusing different relation graphs are effective for improving the ability in emotion recognition. Furthermore, the emotion recognition accuracy of 3-class and 4-class is higher than that of other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

How to determine the scale parameter and the cluster number are two important open issues of spectral clustering remained to be studied. In this paper, it is aimed to overcome these two problems. Firstly, we analyze the principle of spectral clustering from normalized cut. Secondly, on one hand, a weighted local scale was proposed to improve both the classification performance and robustness. On the other hand, we proposed an automatic cluster number estimation method from standpoint of Eigenvectors of its affinity matrix. Finally, a framework of robust and smart spectral clustering method was concluded; it is robust enough to deal with arbitrary distributed datasets and smart enough to estimate cluster number automatically. The proposed method was tested both on artificial datasets and UCI datasets, and experiments prove its availability. [ABSTRACT FROM AUTHOR]

Feature selection and subspace learning are two primary methods to achieve data dimensionality reduction and discriminability enhancement. However, data label information is unavailable in unsupervised learning to guide the dimensionality reduction process. To this end, we propose a soft label guided unsupervised discriminative sparse subspace feature selection (UDS 2 FS) model in this paper, which consists of two superiorities in comparison with the existing studies. On the one hand, UDS 2 FS aims to find a discriminative subspace to simultaneously maximize the between-class data scatter and minimize the within-class scatter. On the other hand, UDS 2 FS estimates the data label information in the learned subspace, which further serves as the soft labels to guide the discriminative subspace learning process. Moreover, the ℓ 2 , 0 -norm is imposed to achieve row sparsity of the subspace projection matrix, which is parameter-free and more stable compared to the ℓ 2 , 1 -norm. Experimental studies to evaluate the performance of UDS 2 FS are performed from three aspects, i.e., a synthetic data set to check its iterative optimization process, several toy data sets to visualize the feature selection effect, and some benchmark data sets to examine the clustering performance of UDS 2 FS. From the obtained results, UDS 2 FS exhibits competitive performance in joint subspace learning and feature selection in comparison with some related models. [ABSTRACT FROM AUTHOR]

Objective: Stroke is the leading cause of disability worldwide. Traditionally, doctors assess stroke rehabilitation assessment, which can be subjective. Therefore, an objective assessment method is required. Methods: In this context, we investigated the changes in brain functional connectivity patterns and corticomuscular coupling in stroke patients during rehabilitation. In this study, electroencephalogram (EEG) and electromyogram (EMG) of stroke patients were collected synchronously at baseline(BL), two weeks after BL, and four weeks after BL. A brain functional network was established, and the corticomuscular coupling relationship was calculated using phase transfer entropy (PTE). Results: We found that during the rehabilitation of stroke patients, the overall connection of the brain functional network was strengthened, and the network characteristic value increased. The average corticomuscular PTE appeared to first decrease and subsequently increase, and the PTE increase in the frontal lobe was significant. Value: In this study, PTE was used for the first time to analyze the relationship between EEG signals in patients with hemiplegia. We believe that our findings contribute to evaluating the rehabilitation of stroke patients with hemiplegia. [ABSTRACT FROM AUTHOR]

About 50-80% of total energy is consumed by signaling in neural networks. A neural network consumes much energy if there are many active neurons in the network. If there are few active neurons in a neural network, the network consumes very little energy. The ratio of active neurons to all neurons of a neural network, that is, the sparseness, affects the energy consumption of a neural network. Laughlin's studies show that the sparseness of an energy-efficient code depends on the balance between signaling and fixed costs. Laughlin did not give an exact ratio of signaling to fixed costs, nor did they give the ratio of active neurons to all neurons in most energy-efficient neural networks. In this paper, we calculated the ratio of signaling costs to fixed costs by the data from physiology experiments. The ratio of signaling costs to fixed costs is between 1.3 and 2.1. We calculated the ratio of active neurons to all neurons in most energy-efficient neural networks. The ratio of active neurons to all neurons in neural networks is between 0.3 and 0.4. Our results are consistent with the data from many relevant physiological experiments, indicating that the model used in this paper may meet neural coding under real conditions. The calculation results of this paper may be helpful to the study of neural coding. [ABSTRACT FROM AUTHOR]

Subspace learning aims to obtain the corresponding low-dimensional representation of high dimensional data in order to facilitate the subsequent data storage and processing. Graph-based subspace learning is a kind of effective subspace learning methods by modeling the data manifold with a graph, which can be included in the general spectral regression (SR) framework. By using the least square regression form as objective function, spectral regression mathematically avoids performing eign-decomposition on dense matrices and has excellent flexibility. Recently, spectral regression has obtained promising performance in diverse applications; however, it did not take the underlying classes/tasks correlation patterns of data into consideration. In this paper, we propose to improve the performance of spectral regression by exploring the correlation among classes with low-rank modeling. The newly formulated low-rank spectral regression (LRSR) model is achieved by decomposing the projection matrix in SR by two factor matrices which were respectively regularized. The LRSR objective function can be handled by the alternating direction optimization framework. Besides some analysis on the differences between LRSR and existing related models, we conduct extensive experiments by comparing LRSR with its full rank counterpart on benchmark data sets and the results demonstrate its superiority. [ABSTRACT FROM AUTHOR]

The Turing Test is a method of testing whether a machine has human intelligence. A novel brain–computer interface (BCI) Turing Test was proposed in the BCI Controlled Robot Contest in World Robot Contest 2022. Contestants developed algorithms that can distinguish if an instruction is issued by a human. Participants collaborated with an electroencephalogram-based BCI to play a soccer game in a virtual scenario. Participants were asked to perform steady-state visual evoked potential (SSVEP) tasks or motor imagery (MI) tasks to control the robots or be in an idle state to mimic the system giving instructions on behalf of the participants. Several algorithms proposed in this competition are developed based on the concept that the idle state is a category in multiclass classification problems. This paper details the algorithms of the top five teams with the best performance in the final, lists the popular classification models and algorithms for MI and SSVEP, and discusses the effectiveness of each approach in improving classification performance and reducing the computation time. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has gradually become a prevailing way in EEG-based emotion recognition research because it can extract features and classify emotions automatically. To fully exploit the underlying information in EEG signals, we propose an emotion recognition method based on cascaded convolutional recurrent neural networks. Firstly, the differential entropy features of each channel signal are transformed into four-dimensional structure data, which are able to contain temporal-spatial-frequency information integratively. Then, the cascaded VGG16 and long short-term memory (LSTM) networks are applied to learn the spatiotemporal information of the samples, and the hidden layer of the last node of LSTM is output to a linear transformation classifier to perform classification. On DEAP dataset, the proposed method gives out an average accuracy of 94.43% and 94.85% in arousal-based and valence-based classification, respectively. On SEED dataset, the method achieves average accuracy of 94.16%. Compared with the existing methods, our method demonstrates superior performances in emotion recognition. [ABSTRACT FROM AUTHOR]

The visual movement illusion (VMI) is a subjective experience. This illusion is produced by watching the subject's motion video. At the same time, VMI evokes awareness of body ownership. We applied the power spectral density (PSD) matrix and the partial directed correlation (PDC) matrix to build the PPDC matrix for the γ2 band (34–98.5 Hz), combining cerebral cortical and musculomotor cortical complexity and PPDC to quantify the degree of body ownership. Thirty-five healthy subjects were recruited to participate in this experiment. The subjects' electroencephalography (EEG) and surface electromyography (sEMG) data were recorded under resting conditions, observation conditions, illusion conditions, and actual seated front-kick movements. The results show the following: (1) VMI activates the cerebral cortex to some extent; (2) VMI enhances cortical muscle excitability in the rectus femoris and medial vastus muscles; (3) VMI induces a sense of body ownership; (4) the use of PPDC values, fuzzy entropy values of muscles, and fuzzy entropy values of the cerebral cortex can quantify whether VMI induces awareness of body ownership. These results illustrate that PPDC can be used as a biomarker to show that VMI affects changes in the cerebral cortex and as a quantitative tool to show whether body ownership awareness arises. [ABSTRACT FROM AUTHOR]

Background: It is crucial to understand the neural feedback mechanisms and the cognitive decision-making of the brain during the processing of rewards. Here, we report the first attempt for a simultaneous electroencephalography (EEG)–functional magnetic resonance imaging (fMRI) study in a gambling task by utilizing tensor decomposition. Methods: First, the single-subject EEG data are represented as a third-order spectrogram tensor to extract frequency features. Next, the EEG and fMRI data are jointly decomposed into a superposition of multiple sources characterized by space-time-frequency profiles using coupled matrix tensor factorization (CMTF). Finally, graph-structured clustering is used to select the most appropriate model according to four quantitative indices. Results: The results clearly show that not only are the regions of interest (ROIs) found in other literature activated, but also the olfactory cortex and fusiform gyrus which are usually ignored. It is found that regions including the orbitofrontal cortex and insula are activated for both winning and losing stimuli. Meanwhile, regions such as the superior orbital frontal gyrus and anterior cingulate cortex are activated upon winning stimuli, whereas the inferior frontal gyrus, cingulate cortex, and medial superior frontal gyrus are activated upon losing stimuli. Conclusion: This work sheds light on the reward-processing progress, provides a deeper understanding of brain function, and opens a new avenue in the investigation of neurovascular coupling via CMTF. [ABSTRACT FROM AUTHOR]

Electroencephalogram (EEG) signals are derived from the central nervous system and inherently difficult to camouflage, leading to the recent popularity of EEG-based emotion recognition. However, due to the non-stationary nature of EEG, inter-subject variabilities become obstacles for recognition models to well adapt to different subjects. In this paper, we propose a novel approach called semi-supervised bipartite graph construction with active EEG sample selection (SBGASS) for cross-subject emotion recognition, which offers two significant advantages. Firstly, SBGASS adaptively learns a bipartite graph to characterize the underlying relationships between labeled and unlabeled EEG samples, effectively implementing the semantic connection for samples from different subjects. Secondly, we employ active sample selection technique in this paper to reduce the impact of negative samples (outliers or noise in the data) on bipartite graph construction. Drawing from the experimental results with the SEED-IV data set, we have gained the following three insights. (1) SBGASS actively rejects negative labeled samples, which helps mitigate the impact of negative samples when constructing the optimal bipartite graph and improves the model performance. (2) Through the learned optimal bipartite graph in SBGASS, the transferability of labeled EEG samples is quantitatively analyzed, which exhibits a decreasing tendency as the distance between each labeled sample and the corresponding class centroid increases. (3) Besides the improved recognition accuracy, the spatial-frequency patterns in emotion recognition are investigated by the acquired projection matrix.Graphical abstract: Electroencephalogram (EEG) signals are derived from the central nervous system and inherently difficult to camouflage, leading to the recent popularity of EEG-based emotion recognition. However, due to the non-stationary nature of EEG, inter-subject variabilities become obstacles for recognition models to well adapt to different subjects. In this paper, we propose a novel approach called semi-supervised bipartite graph construction with active EEG sample selection (SBGASS) for cross-subject emotion recognition, which offers two significant advantages. Firstly, SBGASS adaptively learns a bipartite graph to characterize the underlying relationships between labeled and unlabeled EEG samples, effectively implementing the semantic connection for samples from different subjects. Secondly, we employ active sample selection technique in this paper to reduce the impact of negative samples (outliers or noise in the data) on bipartite graph construction. Drawing from the experimental results with the SEED-IV data set, we have gained the following three insights. (1) SBGASS actively rejects negative labeled samples, which helps mitigate the impact of negative samples when constructing the optimal bipartite graph and improves the model performance. (2) Through the learned optimal bipartite graph in SBGASS, the transferability of labeled EEG samples is quantitatively analyzed, which exhibits a decreasing tendency as the distance between each labeled sample and the corresponding class centroid increases. (3) Besides the improved recognition accuracy, the spatial-frequency patterns in emotion recognition are investigated by the acquired projection matrix. [ABSTRACT FROM AUTHOR]

Neuroelectric signals recorded by micro-electrodes reflect the spontaneous and rhythmic activities of brain neurons. Numerous deep learning frameworks have been designed for various neuroelectric signal decoding tasks, most of which are based on convolutional neural network (CNN) and recurrent neural network (RNN). However, neither CNNs or RNNs can perceive the global dependencies of neural activities in both time and channel dimensions. To address this issue, this paper presents a temporal-channel cascaded transformer network to decode the neural activities of imagined handwriting movements, which can perform imagined handwriting character recognition from spiking activity recorded by two micro-electrode arrays (MEAs). Specifically, we design a temporal-channel cascaded framework and a dense residual transformer encoder structure, which can promote the hierarchical learning and fusion of the temporal and channel features. In addition, a mutual learning strategy of multiple class tokens is proposed to improve classification performance. We conduct performance evaluation experiments on a single-character handwriting-imagination dataset and a sentence handwriting-imagination dataset, which are collected from the public Handwriting BCI dataset. The comparison results demonstrate the superiority of the proposed framework and strategy. Especially in the imagined single-character recognition task, the recognition accuracy of our model can achieve 95.78%, which provides an improvement of + 2 % over the existing state of the art models. [ABSTRACT FROM AUTHOR]

Electroencephalography (EEG) signal classification is a crucial part in motor imagery brain–computer interface (BCI) system. Traditional supervised learning methods have performed well pleasing in EEG classification. Unfortunately, the unlabeled samples are easier to collect than labeled samples. In addition, recent studies have shown that it may degenerate performance of semi-supervised learning by exploiting unlabeled samples without selection. To address these issues, a novel semi-supervised broad learning system with transfer learning (TSS-BLS) is proposed in this paper. First, the pseudo-labels of unlabeled samples are obtained using the joint distribution adaptation algorithm. TSS-BLS is then constructed by an improved manifold regularization framework containing both labeled and pseudo-label information. Finally, the effectiveness of the proposed TSS-BLS is evaluated on three BCI competition datasets and four benchmark datasets from UCI repository and compared with seven state-of-the-art algorithms, including ELM, SS-ELM, HELM, SVM, LapSVM, BLS and GSS-BLS. Experimental results show that the performance of TSS-BLS is superior to BLS and GSS-BLS on average. It is thereby shown that TSS-BLS is safe and efficient for EEG classification. [ABSTRACT FROM AUTHOR]

Concept factorization (CF) is an effective matrix factorization model which has been widely used in many applications. In CF, the linear combination of data points serves as the dictionary based on which CF can be performed in both the original feature space as well as the reproducible kernel Hilbert space (RKHS). The conventional CF treats each dimension of the feature vector equally during the data reconstruction process, which might violate the common sense that different features have different discriminative abilities and therefore contribute differently in pattern recognition. In this paper, we introduce an auto-weighting variable into the conventional CF objective function to adaptively learn the corresponding contributions of different features and propose a new model termed Auto-Weighted Concept Factorization (AWCF). In AWCF, on one hand, the feature importance can be quantitatively measured by the auto-weighting variable in which the features with better discriminative abilities are assigned larger weights; on the other hand, we can obtain more efficient data representation to depict its semantic information. The detailed optimization procedure to AWCF objective function is derived whose complexity and convergence are also analyzed. Experiments are conducted on both synthetic and representative benchmark data sets and the clustering results demonstrate the effectiveness of AWCF in comparison with some related models. [ABSTRACT FROM AUTHOR]

• Simulated microgravity disturbed neural oscillations in the hippocampus of mice. • rTMS treatment significantly modulated theta oscillatory patterns in Hu mice. • rTMS treatment could ameliorate the disturbance of oscillatory patterns in Hu mice. • A potential underlying mechanism was associated with improving the E/I balance. Microgravity, as a part of the stress of space flight, has several negative effects on cognitive functions. Repetitive transcranial magnetic stimulation (rTMS), as a novel non-invasive technique, could be an effective approach to alleviated cognitive decline, applied in both preclinical and clinical studies. Neural oscillations and their interactions are involved in cognitive functions and support the communication of neural information. The neural oscillation could be a window from which we may understand what happens in the brain. The current study aimed to explore if 15 Hz rTMS plays a neural modulation role in a mouse model of hindlimb unloading. We hypothezed that rTMS can improve the cognitive and neural oscillatory deficits induced by hindlimb unloading via maintaining the balance between glutamatergic and GABAergic systems. Our data show that rTMS can significantly alleviate behavior deficits, modulate theta oscillation, improve the disturbed power distribution of theta oscillation and the decreased strength of Cross-Frequency Coupling in the dentate gyrus region, and effectively mitigated the blocked communication of neural information in the perforant pathway (PP)―dentate gyrus (DG) neural pathway in Hu mice. Furthermore, biochemical analysis using high-performance liquid chromatography and Western blot assay confirmed that rTMS increases the low expression of glutamate (Glu) and N-Methyl d -Aspartate receptor subtype 2B (NR2B) and decreases the high expression of γ-aminobutyric acid (GABA), 67 KDa isoform of glutamate decarboxylase (GAD67), and GABA type A receptor subunit alpha1 (GABAAR α1) in the hippocampus of Hu mice. Taken together, the results suggest that rTMS plays a significant neural modulation role in the hippocampal neural activity disorders induced by Hu, which possibly depends on rTMS maintaining the balance of glutamatergic and gamma-aminobutyric acidergic (GABAergic) systems. [ABSTRACT FROM AUTHOR]

We injected mice with an adenovirus coated with an interference sequence, and the results showed that GABAergic synapses signaling pathway, Glutamatergic synapses signaling pathway and Cholinergic synapses pathway were damaged.At the mesoscopic level, the phase synchronization and phase amplitude coupling of the oscillating signals and synaptic plasticity in the hippocampus of mice were all impaired. At the macro level, the cognitive memory ability and the maintenance of working memory of mice were abnormal. Neural oscillation pattern provides a potential diagnosis approach for mental disorders. [Display omitted] • Knocking down Disc1 gene is a proper animal model of mental disorders. • Disc1 gene knockdown mice show memory deficits and synaptic plasticity impairments. • Disc1 gene knockdown mice exhibit abnormal neural oscillations in the hippocampus. • Neural oscillation analysis offers a potential diagnosis method for mental disorders. The gene of Disrupted-in-schizophrenia 1 (Disc1) is closely related to mental diseases with cognitive deficits, but there are few studies on the changes in neural oscillations and recognition memory. Neural oscillations plays a key role in the nervous system in a dynamic form, which is closely related to advanced cognitive activities such as information processing and memory consolidation. Hence, we aimed to investigate if Disc1 knockdown disrupted the normal pattern of neural activities in the mouse hippocampus network, and determined if quantitative neural oscillation approach could be a potential diagnostic tool for mental disorders. In the study, we reported that Disc1 gene, downregulated by short-hairpin RNA (shRNA), not only induced anxiety-like behavior and sociability impairment but also damaged both synaptic plasticity and recognition memory in mice. Moreover, Disc1 knockdown mice exhibited evidently abnormal power spectral distributions, reduced phase synchronizations, and decreased phase-amplitude coupling strength compared to that of normal animals. In addition, transcriptome analyses showed that there were clearly transcriptional changes in Disc1 knockdown mice. Altogether, our findings suggest that the abnormal pattern of neural activities in the hippocampus network disrupts information processing and finally leads to the impairments of synaptic plasticity and recognition in Disc1 knockdown mice, which are possibly associated with the obstruction of neurotransmitter transmission. Importantly, the data imply that the analysis of neural oscillation pattern provides a potential diagnosis approach for mental disorders. [ABSTRACT FROM AUTHOR]

Aiming at the current problem of insufficient extraction of small retinal blood vessels, we propose a retinal blood vessel segmentation algorithm that combines supervised learning and unsupervised learning algorithms. In this study, we use a multiscale matched filter with vessel enhancement capability and a U-Net model with a coding and decoding network structure. Three channels are used to extract vessel features separately, and finally, the segmentation results of the three channels are merged. The algorithm proposed in this paper has been verified and evaluated on the DRIVE, STARE, and CHASE_DB1 datasets. The experimental results show that the proposed algorithm can segment small blood vessels better than most other methods. We conclude that our algorithm has reached 0.8745, 0.8903, and 0.8916 on the three datasets in the sensitivity metric, respectively, which is nearly 0.1 higher than other existing methods. [ABSTRACT FROM AUTHOR]

Emotion recognition has a wide range of potential applications in the real world. Among the emotion recognition data sources, electroencephalography (EEG) signals can record the neural activities across the human brain, providing us a reliable way to recognize the emotional states. Most of existing EEG-based emotion recognition studies directly concatenated features extracted from all EEG frequency bands for emotion classification. This way assumes that all frequency bands share the same importance by default; however, it cannot always obtain the optimal performance. In this paper, we present a novel multi-scale frequency bands ensemble learning (MSFBEL) method to perform emotion recognition from EEG signals. Concretely, we first re-organize all frequency bands into several local scales and one global scale. Then we train a base classifier on each scale. Finally we fuse the results of all scales by designing an adaptive weight learning method which automatically assigns larger weights to more important scales to further improve the performance. The proposed method is validated on two public data sets. For the "SEED IV" data set, MSFBEL achieves average accuracies of 82.75%, 87.87%, and 78.27% on the three sessions under the within-session experimental paradigm. For the "DEAP" data set, it obtains average accuracy of 74.22% for four-category classification under 5-fold cross validation. The experimental results demonstrate that the scale of frequency bands influences the emotion recognition rate, while the global scale that directly concatenating all frequency bands cannot always guarantee to obtain the best emotion recognition performance. Different scales provide complementary information to each other, and the proposed adaptive weight learning method can effectively fuse them to further enhance the performance. [ABSTRACT FROM AUTHOR]

Subspace learning represents a group of algorithms to project high dimensional data onto a low dimensional subspace in order to retain the desirable properties and simultaneously reduce the dimensionality. In graph-based subspace learning algorithms, the quality of graph affects the performance of projection matrix learning a lot. Especially when data is noisy or even grossly corrupted, we cannot guarantee that the constructed graph can accurately depict the inner structure of data. Additionally, the widely used two-stage paradigm of learning the projection matrix on a given graph isolates the connection between these two stages. In this paper, we propose a general framework to get rid of these disadvantages by the inspiration of low-rank matrix recovery and spectral regression-based subspace learning. Concretely, by jointly optimizing the objectives of low-rank representation and spectral regression, on one hand we can automatically get the recovered data from noise and the graph affinity matrix and on the other hand we can perform the projection matrix learning. The formulated joint low-rank and subspace learning (JLRSL) framework can be efficiently optimized by the augmented Lagrange multiplier method. We evaluate the performance of JLRSL by conducting extensive experiments on representative benchmark data sets and the results show that the low-rank learning can greatly facilitate the process of subspace learning, leading to robust feature extraction. Moreover, comparison with the state-of-the-arts is performed. [ABSTRACT FROM AUTHOR]

Fatigue driving can easily lead to road traffic accidents and bring great harm to individuals and families. Recently, electroencephalography- (EEG-) based physiological and brain activities for fatigue detection have been increasingly investigated. However, how to find an effective method or model to timely and efficiently detect the mental states of drivers still remains a challenge. In this paper, we combine common spatial pattern (CSP) and propose a light-weighted classifier, LightFD, which is based on gradient boosting framework for EEG mental states identification. The comparable results with traditional classifiers, such as support vector machine (SVM), convolutional neural network (CNN), gated recurrent unit (GRU), and large margin nearest neighbor (LMNN), show that the proposed model could achieve better classification performance, as well as the decision efficiency. Furthermore, we also test and validate that LightFD has better transfer learning performance in EEG classification of driver mental states. In summary, our proposed LightFD classifier has better performance in real-time EEG mental state prediction, and it is expected to have broad application prospects in practical brain-computer interaction (BCI). [ABSTRACT FROM AUTHOR]

• A bidirectional distillation strategy is proposed to realize the information interaction between hybrid networks. • BIHN optimizes the performance of both CogN and ComN in EEG cognitive recognition by using a bidirectional feedback mode. • Experimental results show the superior performance of BIHN in cognitive recognition compared to many SOTA models. • The BIHN architecture is suitable for various hybrid network pairs related to cognitive and computational networks. Background and objective: Extracting cognitive representation and computational representation information simultaneously from electroencephalography (EEG) data and constructing corresponding information interaction models can effectively improve the recognition capability of brain cognitive status. However, due to the huge gap in the interaction between the two types of information, existing studies have yet to consider the advantages of the interaction of both. Methods: This paper introduces a novel architecture named the bidirectional interaction-based hybrid network (BIHN) for EEG cognitive recognition. BIHN consists of two networks: a cognitive-based network named CogN (e.g., graph convolution network, GCN; capsule network, CapsNet) and a computing-based network named ComN (e.g., EEGNet). CogN is responsible for extracting cognitive representation features from EEG data, while ComN is responsible for extracting computational representation features. Additionally, a bidirectional distillation-based coadaptation (BDC) algorithm is proposed to facilitate information interaction between CogN and ComN to realize the coadaptation of the two networks through bidirectional closed-loop feedback. Results: Cross-subject cognitive recognition experiments were performed on the Fatigue-Awake EEG dataset (FAAD, 2-class classification) and SEED dataset (3-class classification), and hybrid network pairs of GCN + EEGNet and CapsNet + EEGNet were verified. The proposed method achieved average accuracies of 78.76% (GCN + EEGNet) and 77.58% (CapsNet + EEGNet) on FAAD and 55.38% (GCN + EEGNet) and 55.10% (CapsNet + EEGNet) on SEED, outperforming the hybrid networks without the bidirectional interaction strategy. Conclusions: Experimental results show that BIHN can achieve superior performance on two EEG datasets and enhance the ability of both CogN and ComN in EEG processing as well as cognitive recognition. We also validated its effectiveness with different hybrid network pairs. The proposed method could greatly promote the development of brain-computer collaborative intelligence. [ABSTRACT FROM AUTHOR]

A common but easily overlooked affective overlap problem has not been received enough attention in electroencephalogram (EEG)-based emotion recognition research. In real life, affective overlap refers to the current emotional state of human being is sometimes influenced easily by his/her historical mood. In stimulus-evoked EEG collection experiment, due to the short rest interval in consecutive trials, the inner mechanisms of neural responses make subjects cannot switch their emotion state easily and quickly, which might lead to the affective overlap. For example, we might be still in sad state to some extent even if we are watching a comedy because we just saw a tragedy before. In pattern recognition, affective overlap usually means that there exists the feature-label inconsistency in EEG data. To alleviate the impact of inconsistent EEG data, we introduce a variable to adaptively explore the sample inconsistency in emotion recognition model development. Then, we propose a semi-supervised emotion recognition model for joint sample inconsistency and feature importance exploration (SIFIAE). Accordingly, an efficient optimization method to SIFIAE model is proposed. Extensive experiments on the SEED-V dataset demonstrate the effectiveness of SIFIAE. Specifically, SIFIAE achieves 69.10%, 67.01%, 71.50%, 73.26%, 72.07% and 71.35% average accuracies in six cross-session emotion recognition tasks. The results illustrated that the sample weights have a rising trend in the beginning of most trials, which coincides with the affective overlap hypothesis. The feature importance factor indicated the critical bands and channels are more obvious compared with some models without considering EEG feature-label inconsistency. • Affective overlap problem was considered in emotion recognition. • Feature-label inconsistency and feature importance were utilized jointly. • SIFIAE obtained a significant improvement in cross-session emotion recognition. • Obtained sample inconsistency demonstrated affective overlap hypothesis. • Affective activation pattern could be adaptively obtained in our model. [ABSTRACT FROM AUTHOR]

Due to its advantages of high temporal and spatial resolution, the technology of simultaneous electroencephalogram-functional magnetic resonance imaging (EEG-fMRI) acquisition and analysis has attracted much attention, and has been widely used in various research fields of brain science. However, during the fMRI of the brain, ballistocardiogram (BCG) artifacts can seriously contaminate the EEG. As an unpaired problem, BCG artifact removal now remains a considerable challenge. Aiming to provide a solution, this paper proposed a novel modular generative adversarial network (GAN) and corresponding training strategy to improve the network performance by optimizing the parameters of each module. In this manner, we hope to improve the local representation ability of the network model, thereby improving its overall performance and obtaining a reliable generator for BCG artifact removal. Moreover, the proposed method does not rely on additional reference signal or complex hardware equipment. Experimental results show that, compared with multiple methods, the technique presented in this paper can remove the BCG artifact more effectively while retaining essential EEG information. • A novel GAN-based model is designed (BCGGAN) to remove the BCG artifact in simultaneous EEG-fMRI. • A modular training strategy is proposed to optimize the generator network in the BCGGAN model. • The proposed method does not require additional hardware or reference signal, such as carbon fiber sling or ECG signals. • The proposed method can remove the BCG artifact more effectively while retaining useful physiological information. [ABSTRACT FROM AUTHOR]

A brain–computer interface (BCI) based on motor imagery (MI) translates the subject’s motor intention into a control signal through classifying electroencephalogram (EEG) patterns of different imagination tasks, for example, hand movements. Auto-regression (AR) model is one of the popular methods to describe motor imagery patterns, which is widely used by researchers to resolve subject’s motor intention. In this paper, we use joint regression (JR) model and propose an algorithm by combining the coefficients of JR model and spectral powers at two specific frequencies to classify different MI patterns. The algorithm produces a classification accuracy of 90 % on the training data of one subject from BCI2003 Data set III and 80 % on the test data. The results are better than that by using AR model. We also apply the algorithm to MI tasks of one subject in our laboratory, and the classification accuracy can reach 97.86 % on the test data. The results demonstrate that the combination of JR model and spectral powers can achieve much higher accuracy for classification of MI tasks. [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]

Emotions play a crucial role in human communication and affect all aspects of human life. However, to date, there have been few studies conducted on how movements under different emotions influence human brain activity and cortico-muscular coupling (CMC). In this study, for the first time, electroencephalogram (EEG) and electromyogram physiological electrical signals were used to explore this relationship. We performed frequency domain and nonlinear dynamics analyses on EEG signals and used transfer entropy to explore the CMC associated with the emotion-movement relationship. To study the transmission of information between different brain regions, we also constructed a functional brain network and calculated various network metrics using graph theory. We found that, compared with a neutral emotional state, movements made during happy and sad emotions had increased CMC strength and EEG power and complexity. The functional brain network metrics of these three emotional states were also different. Much of the emotion-movement relationship research has been based on subjective expression and external performance. Our research method, however, focused on the processing of physiological electrical signals, which contain a wealth of information and can objectively reveal the inner mechanisms of the emotion-movement relationship. Different emotional states can have a significant influence on human movement. This study presents a detailed introduction to brain activity and CMC. • Combine EEG and EMG to study the relationship between emotion and movement. • Movement associated with happy emotions can increase the complexity of the left hemisphere. • Movement associated with sad emotions can increase the complexity of the right hemisphere. • Increased grip strength leads to increased the CMC strength of EEG→EMG. • The network topology of movement under different emotions is significantly different. [ABSTRACT FROM AUTHOR]

Sparse representation-based classification (SRC) has more advantages in motor imagery EEG pattern recognition, and the quality of dictionary construction directly determines the performance of SRC. In this paper, we proposed a two-dimensional dictionary optimization (TDDO) method to directly improve the performance of SRC. Firstly, an initial dictionary was constructed with multi-band features extracted by filter band common spatial pattern (FBCSP). Then Lasso regression is used to select significant features in each atom synchronously in the horizontal direction, and the KNN-based method is used to clean up noise atoms in the vertical direction. Finally, an SRC method by training samples linearly representing test samples was implemented in classification. The results show the necessity and rationality of TDDO-SRC method. The highest average classification accuracy of 86.5% and 92.4% is obtained on two public datasets. The proposed method has more superior classification accuracy compared to traditional methods and existing winners' methods. The quality of dictionary construction has a great impact on the robustness of SRC. And compared with the original SRC, the classification accuracy of the optimized TDDO-SRC is greatly improved. • The selected features are distributed in frequency bands relevant to the MI tasks. • The atoms cleaning method satisfies the principle of dictionary construction. • A novel two-dimensional dictionary optimization algorithm is proposed. • Compared with SRC, TDDO-SRC significantly improves the classification accuracy. [ABSTRACT FROM AUTHOR]

• Transfer learning can reduce the sample calibration time. • Target alignment make the distribution of samples in source and target domains more similar. • Combining Sub-band filtering and target alignment obtain richer frequency information. • The classification performance of this method is optimal. In the brain computer interface (BCI) field, using sub-band common spatial pattern (SBCSP) and filter bank common spatial pattern (FBCSP) can improve the accuracy of classification by selection a specific frequency band. However, in the cross-subject classification, due to the individual differences between different subjects, the performance is limited. This paper introduces the idea of transfer learning and presents the sub-band target alignment common spatial pattern (SBTACSP) method and applies it to the cross-subject classification of motor imagery (MI) EEG signals. First, the EEG signals are bandpass-filtered into multiple frequency bands (sub-band filtering). Subsequently, the source domain trails are aligned into the target domain space in each frequency band. The CSP algorithm is then employed to extract features among which more representative features are selected by the minimum redundancy maximum relevance (mRMR) approach from each sub-band. Then the features of all sub-bands are fused. Finally, conventional linear discriminant analysis (LDA) algorithm is used for MI classification. Our method is evaluated on Datasets Ⅱa and Ⅱb of the BCI Competition Ⅳ. Compared with six state-of-the-art algorithms, the proposed SBTACSP method performed relatively the best and achieved a mean classification accuracy of 75.15% and 66.85% in cross-subject classification of Datasets Ⅱa and Ⅱb respectively. Therefore, the combination of sub-band filtering and transfer learning achieves superior classification performance compared to either one. The proposed algorithms will greatly promote the practical application of MI based BCIs. [ABSTRACT FROM AUTHOR]

• A new experimental paradigm for exploring the different effects of anode tDCS on M1 and SMA. • The tDCS's effect on the brain-SMA is more obvious in the motor preparation stage. • The effect of tDCS is more obvious in the execution of the entire motor imagination task, but not in the motor preparation stage. • The effect of tDCS on the motor area of the brain is significant, especially in the M1. Related experiments have shown that transcranial direct current stimulation (tDCS) anodal stimulation of the brain's primary motor cortex (M1) and supplementary motor area (SMA) can improve the motor control and clinical manifestations of stroke patients with aphasia and dyskinesia. In this study, to explore the different effects of tDCS on the M1 and SMA in motor imagery, 35 healthy volunteers participated in a double-blind randomized controlled experiment. Five subjects underwent sham stimulation (control), 15 subjects underwent tDCS anode stimulation of the M1, and the remaining 15 subjects underwent tDCS anode stimulation of the SMA. The electroencephalogram data of the subjects' left- and right-hand motor imagery under different stimulation paradigms were recorded. We used a functional brain network and sample entropy to examine the different complexities and functional connectivities in subjects undergoing sham-tDCS and the two stimulation paradigms. The results show that tDCS anodal stimulation of the SMA produces less obvious differences in the motor preparation phase, while tDCS anodal stimulation of the M1 produces significant differences during the motor imaging task execution phase. The effect of tDCS on the motor area of the brain is significant, especially in the M1. [ABSTRACT FROM AUTHOR]

Fatigued driving is one of the main causes of traffic accidents. The electroencephalogram (EEG)-based mental state analysis method is an effective and objective way of detecting fatigue. However, as EEG shows significant differences across subjects, effectively "transfering" the EEG analysis model of the existing subjects to the EEG signals of other subjects is still a challenge. Domain-Adversarial Neural Network (DANN) has excellent performance in transfer learning, especially in the fields of document analysis and image recognition, but has not been applied directly in EEG-based cross-subject fatigue detection. In this paper, we present a DANN-based model, Generative-DANN (GDANN), which combines Generative Adversarial Networks (GAN) to enhance the ability by addressing the issue of different distribution of EEG across subjects. The comparative results show that in the analysis of cross-subject tasks, GDANN has a higher average accuracy of 91.63% in fatigue detection across subjects than those of traditional classification models, which is expected to have much broader application prospects in practical brain–computer interaction (BCI). [ABSTRACT FROM AUTHOR]

Sleep staging is important in sleep research since it is the basis for sleep evaluation and disease diagnosis. Related works have acquired many desirable outcomes. However, most of current studies focus on time-domain or frequency-domain measures as classification features using single or very few channels, which only obtain the local features but ignore the global information exchanging between different brain regions. Meanwhile, brain functional connectivity is considered to be closely related to brain activity and can be used to study the interaction relationship between brain areas. To explore the electroencephalography (EEG)-based brain mechanisms of sleep stages through functional connectivity, especially from different frequency bands, we applied phase-locked value (PLV) to build the functional connectivity network and analyze the brain interaction during sleep stages for different frequency bands. Then, we performed the feature-level, decision-level and hybrid fusion methods to discuss the performance of different frequency bands for sleep stages. The results show that (1) PLV increases in the lower frequency band (delta and alpha bands) and vice versa during different stages of non-rapid eye movement (NREM); (2) alpha band shows a better discriminative ability for sleeping stages; (3) the classification accuracy of feature-level fusion (six frequency bands) reaches 96.91% and 96.14% for intra-subject and inter-subjects respectively, which outperforms decision-level and hybrid fusion methods. [ABSTRACT FROM AUTHOR]

Electroencephalogram (EEG) is an effective indicator for the detection of driver fatigue. Due to the significant differences in EEG signals across subjects, and difficulty in collecting sufficient EEG samples for analysis during driving, detecting fatigue across subjects through using EEG signals remains a challenge. EasyTL is a kind of transfer-learning model, which has demonstrated better performance in the field of image recognition, but not yet been applied in cross-subject EEG-based applications. In this paper, we propose an improved EasyTL-based classifier, the InstanceEasyTL, to perform EEG-based analysis for cross-subject fatigue mental-state detection. Experimental results show that InstanceEasyTL not only requires less EEG data, but also obtains better performance in accuracy and robustness than EasyTL, as well as existing machine-learning models such as Support Vector Machine (SVM), Transfer Component Analysis (TCA), Geodesic Flow Kernel (GFK), and Domain-adversarial Neural Networks (DANN), etc. [ABSTRACT FROM AUTHOR]

Deep learning (DL) methods have been used increasingly widely, such as in the fields of speech and image recognition. However, how to design an appropriate DL model to accurately and efficiently classify electroencephalogram (EEG) signals is still a challenge, mainly because EEG signals are characterized by significant differences between two different subjects or vary over time within a single subject, non-stability, strong randomness, low signal-to-noise ratio. SincNet is an efficient classifier for speaker recognition, but it has some drawbacks in dealing with EEG signals classification. In this paper, we improve and propose a SincNet-based classifier, SincNet-R, which consists of three convolutional layers, and three deep neural network (DNN) layers. We then make use of SincNet-R to test the classification accuracy and robustness by emotional EEG signals. The comparable results with original SincNet model and other traditional classifiers such as CNN, LSTM and SVM, show that our proposed SincNet-R model has higher classification accuracy and better algorithm robustness. [ABSTRACT FROM AUTHOR]