Investigating abnormal brain network characteristics in schizophrenia can improve our understanding of disease mechanisms and help identify potential intervention targets. Graph learning techniques can capture high-dimensional features of large-scale brain networks and offer an inherent advantage for integrating multimodal data. To better integrate multimodal data and accurately localize network abnormalities associated with the disorder, this study proposes a channel-based graph neural network (C-GNN) model. First, node embedding of brain regions was constructed to capture structural connectivity patterns. Second, a branched attention module was introduced to adaptively identify important brain regions through channel attention. Finally, a graph feature-constraint module was developed to extract salient features by computing difference scores across feature channels. The C-GNN model achieved an accuracy of 84.37% in classifying individuals with schizophrenia. Interpretability analysis revealed key abnormal brain regions (e.g. orbital cortex, temporal fusiform cortex, lingual gyrus) and multimodal metrics (such as cortical thickness and ReHo) that contributed substantially to the classification. These findings offer insights into the underlying neural alterations in schizophrenia and may inform the development of targeted intervention strategies.

The behavioral inhibition system (BIS), mediating responses to punishment cues and avoidance behaviors, is implicated in anxiety. However, the neural dynamics underpinning BIS, particularly regarding the temporal variability of brain network interactions, remain less explored. Using resting-state functional magnetic resonance imaging (rs-fMRI) of 181 healthy adults, this study investigated the association between BIS sensitivity and the temporal variability of functional connectivity within and between functional brain networks. This finding revealed a significant positive correlation between BIS scores and temporal variability, specifically in the connectivity involving subnetworks' sensory somatomotor hand network (SSHN)-ventral attention network (VAN), and sensory somatomotor mouth network (SSMN)-VAN. Notably, the high-BIS sensitivity group exhibited significantly greater temporal variability between VAN and SSMN/SSHN compared to the low-BIS sensitivity group. Furthermore, predicted BIS scores based on network variability showed a strong correlation with actual BIS scores (Pearson's [Formula: see text]). Moreover, significant mediation effects highlighted the bridging role of BIS scores between brain network variability and anxiety scale scores. This enhances the comprehension of the relationship between BIS, anxiety, and brain function, while also offering new insights into the pathogenesis of anxiety.

Motor imagery-based brain-computer interface (MI-BCI) faces a critical challenge in achieving effective spatial-temporal feature modeling while maintaining a compact model parameterization. Herein, a lightweight model was proposed, termed as Dual-Attention-EEGNet (DA-EEGNet), which extends the EEGNet backbone by integrating a channel attention module and a depth attention module to selectively emphasize informative electrodes and temporally discriminative features. Two widely used MI benchmark datasets and three evaluation strategies, i.e. subject-dependent scenario, subject-independent scenario, and dataset-independent classification scenario, were utilized to verify the model's performance. Despite its compact design, DA-EEGNet contains merely 3.97[Formula: see text]k trainable parameters and achieves average classification accuracies of [Formula: see text] and [Formula: see text], outperforming or matching existing deep learning approaches that rely on substantially larger parameter counts. Ablation studies further confirm the complementary contributions of the channel and depth attention modules. In addition, visualization analyses, including temporal attention heatmaps and motor-area topographies, demonstrate that DA-EEGNet captures neurophysiologically meaningful spatial-temporal patterns consistent with MI-related brain activity. These results indicate that DA-EEGNet provides a favorable parameter-accuracy trade-off and serves as an efficient and interpretable baseline for MI-BCI applications.

Previous studies have shown that emotions elicit responses in various physiological systems that underlie emotional experience. However, these studies have focused almost solely on how emotions alter the activity of individual physiological systems. Moreover, recent research has highlighted that the human organism is an integrated entity, wherein physiological systems constantly interact to form a dynamic network. Therefore, in this study, we attempted to evaluate whether the physiological network changes with different emotions. We employed multivariate transfer entropy to assess the interaction between six types of emotion-related physiological signals and subsequently constructed physiological networks across forty emotional trials. Additionally, topological parameters were applied to each physiological network to quantify its overall structure. The results revealed that emotion has a significant impact on the topology of physiological networks, indicating the integrated behavior of physiological systems in response to emotion at the macro-scale level. Furthermore, different physiological systems may play diverse roles within the physiological network. In summary, we propose a system-level integrative approach, referred to as the physiological network, to study the physiological responses to emotion. This not only contributes to a deeper understanding of emotional mechanisms but also demonstrates potential for disease research and engineering applications.

The use of dry electrode electroencephalography (EEG) systems holds significant importance in advancing the everyday application of emotion recognition. However, adapting it to real-world applications faces unique challenges due to low signal-to-noise ratios and unreliable emotion labels. To address these challenges, we propose a Cross-Device Representation Consistency (CDRC) pre-training paradigm for dry EEG emotion recognition, where the self-supervised signal is provided by the distance between representations embedded in wet and dry EEG components and trained via contrastive estimation. Specifically, we employ a dual-branch embedding prediction task coupled with contrastive feature alignment module to extract robust and distinctive features from dry electrode EEG signals. We evaluate our model on an available emotional dataset PaDWEED, extensive experiments demonstrate that CDRC performs comparably to fully supervised training and achieves state-of-the-art results compared to several self-supervised approaches. Moreover, the remarkable performance on subject-independent tasks highlights its effectiveness in addressing and mitigating subject variability.

Emotion recognition based on brain networks has attracted increasing research attention due to its ability to reveal the information interactions between brain regions under different emotional states. However, there are still two challenges in practical applications: 1) The high dimensionality of brain networks can lead to issues of feature redundancy, overfitting, and high computational costs; 2) Electroencephalography (EEG) signals are susceptible to outlier noise, and label noise caused by mismatches between the stimuli labels used to induce emotions and the individuals’ actual emotional responses can significantly impact emotion recognition. To address these challenges, we propose a supervised graph embedding algorithm based on the L1-norm space (L1-SGE). This method leverages the local structure and class information of the original data for discriminative subspace learning, achieving a low-dimensional representation of high-dimensional networks. Additionally, the constraints of the L1-norm space enable the method to effectively suppress outliers and label noise. The performance on publicly available emotional EEG databases has successfully validated the effectiveness of the proposed method in low-dimensional feature representation and noise suppression. Furthermore, this method not only offers a powerful tool for research in affective brain-computer interfaces but also provides a potential solution for pattern recognition tasks facing similar challenges in the field of artificial intelligence.

Real-time emotion recognition provides promising applications for mental healthcare monitoring and human-computer interaction design. Electroencephalography (EEG) emotion recognition has become a hot topic in the field of affective computing and intelligent brain-computer interface (BCI), and it is a feasible solution for achieving real-time emotion recognition. However, due to the uncertainty and individual specificity of emotional cognition, there are still some challenges in achieving efficient online emotion decoding applications. To address this, in this work, we propose an online emotion decoding method named DMSGL (Real-Time EEG Emotion Recognition from Dynamic Mixed Spatiotemporal Graph Learning). Specifically, in the DMSGL, we propose to explore the latent emotion-related graph features from EEG with cognition-inspired and data-driven learning strategies, and the temporal analysis with attention learning is utilized to further extract the robust spatiotemporal graph patterns for efficient EEG emotion decoding. Both simulated online emotion decoding and real-time emotion monitoring experimental results have consistently indicated that the proposed DMSGL can effectively satisfy the application requirements of real-time emotion decoding and achieves an accuracy of 68.35% in real-world online scenarios. Compared with other baseline methods, the proposed DMSGL has improved by 2-5% in the scenario of real-time emotion recognition. In conclusion, the proposed DMSGL provides a promising solution for realizing real-time emotion recognition and further exploring related applications. Our code is released on https://github.com/UESTC-BAC/DMSGL.

Hand gesture recognition (HGR) plays a crucial role in human-computer interaction systems within the Internet of Things (IoT). Recent HGR methods often rely on vision-based images or videos, which are limited in terms of occluded fingers and high computational cost due to complex neural networks. In contrast, wearable sensors like inertial measurement units (IMUs) and flexible sensors can handle hand self-obscuration. However, there are two unresolved issues. First, using a single modality is hard to balance high precision and low latency. Second, existing multimodal-based approaches lack deep inter-modal coupling to effectively address IMU drift and mechanical coupling of flexible sensors. To address these problems, we propose FI-HGR (HGR based on flexible and inertial data). FI-HGR comprises a sensor-integrated data glove and a novel Cascaded Complementary-Stochastic Fusion Algorithm (CS-Algorithm). CS-Algorithm employs six Mahony filters to estimate the state quaternion of each IMU, along with an Extended Kalman Filter that continuously corrects IMU drift based on the index finger’s bending angle sensed by a flexible sensor. This design allows a single flexible sensor to calibrate multiple IMUs and introduces a hard constraint, resolving the sensor drift problems that previous methods cannot. Based on the CS-Algorithm outputs, precise finger bending angles are estimated in real time. Subjective and objective experimental results show that our approach effectively addresses IMU drift and mechanical coupling in flexible sensors, reduces gesture tracking error to approximately 3.4∘, and significantly improves both recognition accuracy and operational efficiency.

Recent research has consistently indicated that the fusion of electroencephalography (EEG) features from multiple modalities can integrate cognitive state expressions across diverse dimensions, resulting in a substantial increase in emotion recognition accuracy. However, redundant information within the fused multimodal features could lead to the curse of dimensionality and overfitting of the learning model. In this work, we propose a multiscale EEG feature fusion and representation strategy for EEG emotion recognition named manifold of multiscale information fusion (MMIF), in which the optimal manifold of the multiscale fusion of local and global brain activation patterns can be automatically learned to realize an efficient representation of emotional EEG signals. To evaluate the performance, in this work, both off- and online EEG emotion recognition experiments were conducted, and the experimental results consistently verified the effectiveness and feasibility of the MMIF applied in real-time emotion decoding systems. Furthermore, the analytical experiments confirmed the discriminative capabilities and cognitive interpretability of the MMIF. In summary, the proposed MMIF model may provide an efficient avenue for exploring representations and enhancing the discrimination of multimodal fusion features, which may also provide a promising solution for designing online affective braincomputer interaction systems.

OBJECTIVE: Freezing of Gait (FOG) significantly impacts daily activities of Parkinson's disease (PD) patients. Despite the potential of wearable sensors in predicting FOG, challenges persist, including the brief prediction interval before FOG onset, limited generalization across patients, and the inconvenience of multiple sensors. Addressing one issue often aggravates others, making it difficult to achieve suitable concurrent solutions to all these challenges. METHODS: We introduce the PhysioGait Predictive Network (PhysioGPN), a deep learning framework designed to predict FOG events in PD patients at least 2 seconds prior to onset. The model architecture incorporates four key strategies: 1) Detection of progressive motion changes using large convolutional kernels; 2) Unraveling the complexity of motion coordination and gait dynamics using multi-dimensional and multi-scale convolution; 3) Capture gait self-similarity and asymmetry with twin-tower structure; 4) Promoting cross-domain information exchange with multi-domain attention. Furthermore, we propose a framework based on knowledge distillation (KD), reducing the model's dependence on multiple sensors while maintaining prediction accuracy. RESULTS: The model achieves an 85.8% Area Under the Curve (AUC) in FOG prediction. When reducing the number of sensors, KD mitigates the decline in performance and increases the AUC by 5.1%, compared to scenarios without KD. CONCLUSION: Our research proposes a practical solution to the challenges of FOG prediction, demonstrating the effectiveness of the KD approach for lightweight wearable sensors in rehabilitation engineering. SIGNIFICANCE: Our findings offer valuable insights for addressing multiple challenges in the practical application of wearable devices.

The goal of the fusion process in RGB-D reconstruction systems is to verify and update the 3D model while ensuring both completeness and accuracy. However, achieving precise dense correspondences in a point-to-point or pixel model during this process is challenging and computationally intensive. To address this challenge, we propose a Manifold Embedding framework that facilitates rapid point-to-surface fusion, removing the need for direct point-to-point or pixel correspondences. Our approach consists of three main steps: 1)Manifold Voxel: We transform discrete point sets into smooth surfaces using the Implicit Moving Least Squares (IMLS) method; 2)Two-Step Filtering: We enhance reconstruction accuracy through a two-step filtering technique that evaluates sampling points based on probabilistic measures; 3)Embedding for Smooth Surface: Lastly, we embed points into a smooth manifold surface represented via IMLS, ensuring high-quality reconstructed surfaces. Extensive experiments on both real and synthetic 3D scenes demonstrate the effectiveness of our Manifold Embedding framework. For instance, on the publicReplicadataset, our method surpasses state-of-the-art fusion techniques regarding both completeness and accuracy. Our average accuracy is 2.11 cm and completeness is 2.80 cm, while NICE-SLAM achieves 2.85 cm and 3.00 cm, respectively (with lower values indicating better performance). Overall, our proposed method provides superior reconstruction quality and enhanced computational efficiency (See Fig. 1).

Spiking neural networks (SNNs) leverage their spike-driven nature to achieve high energy efficiency, positioning them as a promising alternative to traditional artificial neural networks (ANNs). The spiking decoder, a crucial component for output, significantly affects the performance of SNNs. However, current rate coding schemes for decoding of SNNs often lack robustness and do not have a training framework suitable for robust learning, while alternatives to rate coding generally produce worse overall performance. To address these challenges, we propose spatiotemporal prototype (STP) learning for SNNs, which uses multiple learnable binarized prototypes for distance-based decoding. In addition, we introduce a cotraining framework that jointly optimizes prototypes and model parameters, enabling mutual adaptation of the two components. STP learning clusters feature centers through supervised learning to ensure effective aggregation around the prototypes, while maintaining enough spacing between prototypes to handle noise and interference. This dual capability results in superior stability and robustness. On eight benchmark datasets with diverse challenges, the STP-SNN model achieves performance comparable to or superior to state-of-the-art methods. Notably, STP learning demonstrates exceptional robustness and stability in multitask experiments. Overall, these findings reveal that STP learning is an effective means of improving the performance and robustness of SNNs.

Vision transformer (ViT) and its variants have achieved remarkable success in various tasks. The key characteristic of these ViT models is to adopt different aggregation strategies of spatial patch information within the artificial neural networks (ANNs). However, there is still a key lack of unified representation of different ViT architectures for systematic understanding and assessment of model representation performance. Moreover, how those well-performing ViT ANNs are similar to real biological neural networks (BNNs) is largely unexplored. To answer these fundamental questions, we, for the first time, propose a unified and biologically plausible relational graph representation of ViT models. Specifically, the proposed relational graph representation consists of two key subgraphs: an aggregation graph and an affine graph. The former considers ViT tokens as nodes and describes their spatial interaction, while the latter regards network channels as nodes and reflects the information communication between channels. Using this unified relational graph representation, we found that: 1) model performance was closely related to graph measures; 2) the proposed relational graph representation of ViT has high similarity with real BNNs; and 3) there was a further improvement in model performance when training with a superior model to constrain the aggregation graph.

Emotion recognition based on electroencephalogram (EEG) is fundamentally associated with human-like intelligence system. However, due to the noise-sensitive characteristics of EEGs and the individual variability of emotions, it is very challenging to extract inherent emotion dependent patterns from emotional EEG signals. In this work, we propose a L1-norm space defined discriminative brain network manifold learning model (L1-SGL), in which the EEG noise outliers can be effectively separated and the pseudolabeled samples caused by subjective feelings can be automatically corrected. Off-line experimental results consistently indicate that the L1-SGL can effectively suppress the influence of noise and achieve an incomparable superiority performance over other existing methods in EEG emotion recognition. Besides, benefiting from the time efficiency of the L1-SGL, an online emotion monitoring and regulation system is further implemented in this work. On-line emotion decoding experimental results (86.30%) of 25 participants prove that the L1-SGL can effectively satisfy the real-time requirements of on-line emotional monitoring applications, and the significant negative emotion regulation experimental results ( $p \lt 0.001$ ) further confirm the feasibility and effectiveness of L1-SGL model in real-time emotion regulation and interactive applications. Overall, the L1-SGL provides a promising solution for the real-time online affective brain-computer interfaces (aBCIs) and the intelligent clinical closed-loop treatments.

Benefiting from the high-temporal resolution of electroencephalogram (EEG), EEG-based emotion recognition has become one of the hotspots of affective computing. For EEG-based emotion recognition systems, it is crucial to utilize state-of-the-art learning strategies to automatically learn emotion-related brain cognitive patterns from emotional EEG signals, and the learned stable cognitive patterns effectively ensure the robustness of the emotion recognition system. In this work, to realize the efficient decoding of emotional EEG, we propose a graph learning system [Graph Convolutional Network framework with Brain network initial inspiration and Fused attention mechanism (BF-GCN)] inspired by the brain cognitive mechanism to automatically learn graph patterns from emotional EEG and improve the performance of EEG emotion recognition. In the proposed BF-GCN, three graph branches, i.e., cognition-inspired functional graph branch, data-driven graph branch, and fused common graph branch, are first elaborately designed to automatically learn emotional cognitive graph patterns from emotional EEG signals. And then, the attention mechanism is adopted to further capture the brain activation graph patterns that are related to emotion cognition to achieve an efficient representation of emotional EEG signals. Essentially, the proposed BF-CGN model is a cognition-inspired graph learning neural network model, which utilizes the spectral graph filtering theory in the automatic learning and extracting of emotional EEG graph patterns. To evaluate the performance of the BF-GCN graph learning system, we conducted subject-dependent and subject-independent experiments on two public datasets, i.e., SEED and SEED-IV. The proposed BF-GCN graph learning system has achieved 97.44% (SEED) and 89.55% (SEED-IV) in subject-dependent experiments, and the results in subject-independent experiments have achieved 92.72% (SEED) and 82.03% (SEED-IV), respectively. The state-of-the-art performance indicates that the proposed BF-GCN graph learning system has a robust performance in EEG-based emotion recognition, which provides a promising direction for affective computing.

Autism spectrum disorder (ASD) is a mental disorder that severely affects social interaction and communication skills. A timely and accurate diagnosis is crucial for effective intervention. However, as an objective diagnostic tool, the existence of spurious connections and noise within the functional connectivity (FC) matrix in the brain contribute to the complexity of ASD diagnosis. In this study, we propose an unsupervised sparse graph structure learning framework, called NeuroSparse, for reasoning regarding the topological connectivity relationships between brain regions. First, the framework fuses spatiotemporal information from static and dynamic FC to provide a comprehensive representation of complex brain networks. Next, the context and neighborhood information of nodes are captured as their local and global embeddings, and the latent representations of node embeddings are learned, which are then decoded to infer the connectivity relationships among the regions of interest. Finally, we introduce three innovative loss functions to regularize and optimize the sparse representation generation process. The experimental results showed that, after learning using NeuroSparse, the FC matrix achieved state-of-the-art performance in ASD diagnosis using only a simple multilayer perceptron, indicating its wide-ranging application potential. The code is available at https://github.com/yangshuqigit/NeuroSparse.

Cognitive flexibility refers to the capacity to shift between patterns of mental function and relies on functional activity supported by anatomical structures. However, how the brain's structural-functional covarying is preconfigured in the resting state to facilitate cognitive flexibility under tasks remains unrevealed. Herein, we investigated the potential relationship between individual cognitive flexibility performance during the trail-making test (TMT) and structural-functional covariation of the large-scale multimodal covariance network (MCN) using magnetic resonance imaging (MRI) and electroencephalograph (EEG) datasets of 182 healthy participants. Results show that cognitive flexibility correlated significantly with the intra-subnetwork covariation of the visual network (VN) and somatomotor network (SMN) of MCN. Meanwhile, inter-subnetwork interactions across SMN and VN/default mode network/frontoparietal network (FPN), as well as across VN and ventral attention network (VAN)/dorsal attention network (DAN) were also found to be closely related to individual cognitive flexibility. After using resting-state MCN connectivity as representative features to train a multi-layer perceptron prediction model, we achieved a reliable prediction of individual cognitive flexibility performance. Collectively, this work offers new perspectives on the structural-functional coordination of cognitive flexibility and also provides neurobiological markers to predict individual cognitive flexibility.

Idiopathic generalized epilepsy (IGE) is characterized by cryptogenic etiology and the striatum and cerebellum are recognized as modulators of epileptic network. We collected simultaneous electroencephalogram and functional magnetic resonance imaging data from 145 patients with IGE, 34 of whom recorded interictal epileptic discharges (IEDs) during scanning. In states without IEDs, hierarchical connectivity was performed to search core cortical regions which might be potentially modulated by striatum and cerebellum. Node-node and edge-edge moderation models were constructed to depict direct and indirect moderation effects in states with and without IEDs. Patients showed increased hierarchical connectivity with sensorimotor cortices (SMC) and decreased connectivity with regions in the default mode network (DMN). In the state without IEDs, striatum, cerebellum, and thalamus were linked to weaken the interactions of regions in the salience network (SN) with DMN and SMC. In periods with IEDs, overall increased moderation effects on the interaction between regions in SN and DMN, and between regions in DMN and SMC were observed. The thalamus and striatum were implicated in weakening interactions between regions in SN and SMC. The striatum and cerebellum moderated the cortical interaction among DMN, SN, and SMC in alliance with the thalamus, contributing to the dysfunction in states with and without IEDs in IGE. The current work revealed state-specific modulation effects of striatum and cerebellum on thalamocortical circuits and uncovered the potential core cortical targets which might contribute to develop new clinical neuromodulation techniques.

Schizophrenia is accompanied by aberrant interactions of intrinsic brain networks. However, the modulatory effect of electroencephalography (EEG) rhythms on the functional connectivity (FC) in schizophrenia remains unclear. This study aims to provide new insight into network communication in schizophrenia by integrating FC and EEG rhythm information. After collecting simultaneous resting-state EEG-functional magnetic resonance imaging data, the effect of rhythm modulations on FC was explored using what we term "dynamic rhythm information." We also investigated the synergistic relationships among three networks under rhythm modulation conditions, where this relationship presents the coupling between two brain networks with other networks as the center by the rhythm modulation. This study found FC between the thalamus and cortical network regions was rhythm-specific. Further, the effects of the thalamus on the default mode network (DMN) and salience network (SN) were less similar under alpha rhythm modulation in schizophrenia patients than in controls ([Formula: see text]). However, the similarity between the effects of the central executive network (CEN) on the DMN and SN under gamma modulation was greater ([Formula: see text]), and the degree of coupling was negatively correlated with the duration of disease ([Formula: see text], [Formula: see text]). Moreover, schizophrenia patients exhibited less coupling with the thalamus as the center and greater coupling with the CEN as the center. These results indicate that modulations in dynamic rhythms might contribute to the disordered functional interactions seen in schizophrenia.

OBJECTIVE: Cognition is an essential human function, and its development in infancy is crucial. Traditionally, pediatricians used clinical observation or medical imaging to assess infants' current cognitive development (CD) status. The object of pediatricians' greater concern is however their future outcomes, because high-risk infants can be identified early in life for intervention. However, this opportunity has not yet been realized. Fortunately, some recent studies have shown that the general movement (GM) performance of infants around 3-4 months after birth might reflect their future CD status, which gives us an opportunity to achieve this goal by cameras and artificial intelligence. METHODS: First, infants' GM videos were recorded by cameras, from which a series of features reflecting their bilateral movement symmetry (BMS) were extracted. Then, after at least eight months of natural growth, the infants' CD status was evaluated by the Bayley Infant Development Scale, and they were divided into high-risk and low-risk groups. Finally, the BMS features extracted from the early recorded GM videos were fed into the classifiers, using late infant CD risk assessment as the prediction target. RESULTS: The area under the curve, recall and precision values reached 0.830, 0.832, and 0.823 for two-group classification, respectively. CONCLUSION: This pilot study demonstrates that it is possible to automatically predict the CD of infants around the age of one year based on their GMs recorded early in life. SIGNIFICANCE: This study not only helps clinicians better understand infant CD mechanisms, but also provides an economical, portable and non-invasive way to screen infants at high-risk early to facilitate their recovery.

Available evidence suggests that dynamic functional connectivity can capture time-varying abnormalities in brain activity in resting-state cerebral functional magnetic resonance imaging (rs-fMRI) data and has a natural advantage in uncovering mechanisms of abnormal brain activity in schizophrenia (SZ) patients. Hence, an advanced dynamic brain network analysis model called the temporal brain category graph convolutional network (Temporal-BCGCN) was employed. Firstly, a unique dynamic brain network analysis module, DSF-BrainNet, was designed to construct dynamic synchronization features. Subsequently, a revolutionary graph convolution method, TemporalConv, was proposed based on the synchronous temporal properties of features. Finally, the first modular test tool for abnormal hemispherical lateralization in deep learning based on rs-fMRI data, named CategoryPool, was proposed. This study was validated on COBRE and UCLA datasets and achieved 83.62% and 89.71% average accuracies, respectively, outperforming the baseline model and other state-of-the-art methods. The ablation results also demonstrate the advantages of TemporalConv over the traditional edge feature graph convolution approach and the improvement of CategoryPool over the classical graph pooling approach. Interestingly, this study showed that the lower-order perceptual system and higher-order network regions in the left hemisphere are more severely dysfunctional than in the right hemisphere in SZ, reaffirmings the importance of the left medial superior frontal gyrus in SZ. Our code was available at: https://github.com/swfen/Temporal-BCGCN.

Spiking neural networks (SNNs) mimic brain computational strategies, and exhibit substantial capabilities in spatiotemporal information processing. As an essential factor for human perception, visual attention refers to the dynamic process for selecting salient regions in biological vision systems. Although visual attention mechanisms have achieved great success in computer vision applications, they are rarely introduced into SNNs. Inspired by experimental observations on predictive attentional remapping, we propose a new spatial-channel-temporal-fused attention (SCTFA) module that can guide SNNs to efficiently capture underlying target regions by utilizing accumulated historical spatial-channel information in the present study. Through a systematic evaluation on three event stream datasets (DVS Gesture, SL-Animals-DVS, and MNIST-DVS), we demonstrate that the SNN with the SCTFA module (SCTFA-SNN) not only significantly outperforms the baseline SNN (BL-SNN) and two other SNN models with degenerated attention modules, but also achieves competitive accuracy with the existing state-of-the-art (SOTA) methods. Additionally, our detailed analysis shows that the proposed SCTFA-SNN model has strong robustness to noise and outstanding stability when faced with incomplete data, while maintaining acceptable complexity and efficiency. Overall, these findings indicate that incorporating appropriate cognitive mechanisms of the brain may provide a promising approach to elevate the capabilities of SNNs.

Graph neural networks (GNNs) have received increasing interest in the medical imaging field given their powerful graph embedding ability to characterize the non-Euclidean structure of brain networks based on magnetic resonance imaging (MRI) data. However, previous studies are largely node-centralized and ignore edge features for graph classification tasks, resulting in moderate performance of graph classification accuracy. Moreover, the generalizability of GNN model is still far from satisfactory in brain disorder [e.g., autism spectrum disorder (ASD)] identification due to considerable individual differences in symptoms among patients as well as data heterogeneity among different sites. In order to address the above limitations, this study proposes a novel adversarial learning-based node-edge graph attention network (AL-NEGAT) for ASD identification based on multimodal MRI data. First, both node and edge features are modeled based on structural and functional MRI data to leverage complementary brain information and preserved in the constructed weighted adjacent matrix for individuals through the attention mechanism in the proposed NEGAT. Second, two AL methods are employed to improve the generalizability of NEGAT. Finally, a gradient-based saliency map strategy is utilized for model interpretation to identify important brain regions and connections contributing to the classification. Experimental results based on the public Autism Brain Imaging Data Exchange I (ABIDE I) data demonstrate that the proposed framework achieves a classification accuracy of 74.7% between ASD and typical developing (TD) groups based on 1007 subjects across 17 different sites and outperforms the state-of-the-art methods, indicating satisfying classification ability and generalizability of the proposed AL-NEGAT model. Our work provides a powerful tool for brain disorder identification.

The cerebral cortex is folded as gyri and sulci, which provide the foundation to unveil anatomo-functional relationship of brain. Previous studies have extensively demonstrated that gyri and sulci exhibit intrinsic functional difference, which is further supported by morphological, genetic, and structural evidences. Therefore, systematically investigating the gyro-sulcal (G-S) functional difference can help deeply understand the functional mechanism of brain. By integrating functional magnetic resonance imaging (fMRI) with advanced deep learning models, recent studies have unveiled the temporal difference in functional activity between gyri and sulci. However, the potential difference of functional connectivity, which represents functional dependency between gyri and sulci, is much unknown. Moreover, the regularity and variability of the G-S functional connectivity difference across multiple task domains remains to be explored. To address the two concerns, this study developed new anatomy-guided spatio-temporal graph convolutional networks (AG-STGCNs) to investigate the regularity and variability of functional connectivity differences between gyri and sulci across multiple task domains. Based on 830 subjects with seven different task-based and one resting state fMRI (rs-fMRI) datasets from the public Human Connectome Project (HCP), we consistently found that there are significant differences of functional connectivity between gyral and sulcal regions within task domains compared with resting state (RS). Furthermore, there is considerable variability of such functional connectivity and information flow between gyri and sulci across different task domains, which are correlated with individual cognitive behaviors. Our study helps better understand the functional segregation of gyri and sulci within task domains as well as the anatomo-functional-behavioral relationship of the human brain.

Multichannel electroencephalogram (EEG) is an array signal that represents brain neural networks and can be applied to characterize information propagation patterns for different emotional states. To reveal these inherent spatial graph features and increase the stability of emotion recognition, we propose an effective emotion recognition model that performs multicategory emotion recognition with multiple emotion-related spatial network topology patterns (MESNPs) by learning discriminative graph topologies in EEG brain networks. To evaluate the performance of our proposed MESNP model, we conducted single-subject and multisubject four-class classification experiments on two public datasets, MAHNOB-HCI and DEAP. Compared with existing feature extraction methods, the MESNP model significantly enhances the multiclass emotional classification performance in the single-subject and multisubject conditions. To evaluate the online version of the proposed MESNP model, we designed an online emotion monitoring system. We recruited 14 participants to conduct the online emotion decoding experiments. The average online experimental accuracy of the 14 participants was 84.56%, indicating that our model can be applied in affective brain-computer interface (aBCI) systems. The offline and online experimental results demonstrate that the proposed MESNP model effectively captures discriminative graph topology patterns and significantly improves emotion classification performance. Moreover, the proposed MESNP model provides a new scheme for extracting features from strongly coupled array signals.

Electroencephalogram (EEG) brain network embodies the brain’s coordination and interaction mechanism, and the transformations of emotional states are usually accompanied with changes in brain network spatial topologies. To effectively characterize emotions, in this work, we propose a cognition-inspired graph embedding model in the L1-norm space (L1-CGE) to learn an optimal low-dimensional embedded manifold for emotional brain networks. In the L1-CGE, the original brain networks are first encoded in the affinity space with the proposed cognition-inspired metric to construct the latent geometry manifold structure of emotional brain networks, and then the graph learning objective function is defined in the L1-norm space to obtain the optimal low-dimensional representations of brain networks. Essentially, the modularized community structures of emotional brain networks can be effectively emphasized by the L1-CGE to realize an effective depiction for emotions. Compared with existing methods, the L1-CGE model has achieved state-of-the-art performance on three public emotional EEG datasets in off-line conditions. Besides, the robust real-time experimental results have been achieved with the on-line emotion decoding system designed with L1-CGE. Both off- and on-line experimental results consistently demonstrate that the proposed L1-CGE is promising to provide a potential solution for the real-time affective brain-computer interface (aBCI) system.

Cognitive processes induced by the specific task are underpinned by intrinsic anatomical structures with functional neural activation patterns. However, current covariance network analysis still pays much attention to brain morphologies or baseline activity due to the lack of an effective method for capturing the structural-functional covarying during tasks. Here, a multimodal covariance network (MCN) construction method was proposed to identify inter-regional covariations of the structural skeleton and functional activities by simultaneous magnetic resonance imaging and electroencephalogram (EEG). Results from two independent cohorts confirmed that MCNs could capture cognition-specific hierarchical modules in joint comprehensive multimodal features well, especially when time-resolved EEG was further integrated. The quantitative evaluation further demonstrates significantly larger modularity of MCN integrating fine-grained features from EEG. The application to the discovery cohort identified prominent modular covarying across the default mode and salience networks at rest, while the visual oddball task was accomplished by synchronous structural-functional cooperation within networks associated with attention control and working memory updating. Strikingly, the results of an external validation cohort showed a different covariant pattern corresponding to decision-specific cognitive modules. Overall, the results suggested that multimodal covariance analysis provides a reliable definition of multistate neural cognitive networks, further discloses modular-specific structural and functional co-variation.

Previous studies have shown that eye activities, including blinks, can indicate the psychological state of an individual. However, almost all previous studies analyzing blinks merely concentrated on traditional descriptive statistics, which are unable to reflect their dynamic processes. Furthermore, the states of non-blink (opening the eyes) and blink alternate with each other, forming a physiological cycle. If we only investigate blinks alone, it may be inadequate to describe how blinking works. Therefore, we attempted to recognize the affective state (“relaxation” vs. “stress”) of an individual through the dynamics of blink and non-blink cyclicity (BNBC), as one example, to illustrate this method. First, the “Stroop Test” was employed for emotion elicitation. Then, features were extracted from a categorical time series (0: non-blink; 1: blink), which was recorded by the eye-tracking system. Finally, the areas under the receiver operating characteristic curve (AUC) values were obtained via eight commonly used classifiers. The results show that, compared with the traditional approaches for blink analysis, BNBC exhibits more compelling proficiency to detect stress. In summation, BNBC can be considered a new type of psychophysiological measure, which could be widely applied in psychology, medicine, and engineering.

As a kind of biological network, the brain network conduces to understanding the mystery of high-efficiency information processing in the brain, which will provide instructions to develop efficient brain-like neural networks. Large-scale dynamical functional network connectivity (dFNC) provides a more context-sensitive, dynamical, and straightforward sight at a higher network level. Nevertheless, dFNC analysis needs good enough resolution in both temporal and spatial domains, and the construction of dFNC needs to capture the time-varying correlations between two multivariate time series with unmatched spatial dimensions. Effective methods still lack. With well-developed source imaging techniques, electroencephalogram (EEG) has the potential to possess both high temporal and spatial resolutions. Therefore, we proposed to construct the EEG large-scale cortical dFNC based on brain atlas to probe the subtle dynamic activities in the brain and developed a novel method, that is, wavelet coherence-S estimator (WTCS), to assess the dynamic couplings among functional subnetworks with different spatial dimensions. The simulation study demonstrated its robustness and availability of applying to dFNC. The application in real EEG data revealed the appealing "Primary peak" and "P3-like peak" in dFNC network properties and meaningful evolutions in dFNC network topology for P300. Our study brings new insights for probing brain activities at a more dynamical and higher hierarchical level and pushing forward the development of brain-inspired artificial neural networks. The proposed WTCS not only benefits the dFNC studies but also gives a new solution to capture the time-varying couplings between the multivariate time series that is often encountered in signal processing disciplines.

As a novel approach for people to directly communicate with an external device, the study of brain-computer interfaces (BCIs) has become well-rounded. However, similar to the real-world scenario, where individuals are expected to work in groups, the BCI systems should be able to replicate group attributes. We proposed a 4-order cumulants feature extraction method (CUM4-CSP) based on the common spatial patterns (CSP) algorithm. Simulation experiments conducted using motion visual evoked potentials (mVEP) EEG data verified the robustness of the proposed algorithm. In addition, to freely choose paradigms, we adopted the mVEP and steady-state visual evoked potential (SSVEP) paradigms and designed a multimodal collaborative BCI system based on the proposed CUM4-CSP algorithm. The feasibility of the proposed multimodal collaborative system framework was demonstrated using a multiplayer game controlling system that simultaneously facilitates the coordination and competitive control of two users on external devices. To verify the robustness of the proposed scheme, we recruited 30 subjects to conduct online game control experiments, and the results were statistically analyzed. The simulation results prove that the proposed CUM4-CSP algorithm has good noise immunity. The online experimental results indicate that the subjects could reliably perform the game confrontation operation with the selected BCI paradigm. The proposed CUM4-CSP algorithm can effectively extract features from EEG data in a noisy environment. Additionally, the proposed scheme may provide a new solution for EEG-based group BCI research.

Very fast ripples (VFRs, 500-1000 Hz) are considered more specific than high-frequency oscillations (80-500 Hz) as biomarkers of epileptogenic zones. Although VFRs are frequent abnormal phenomena in epileptic seizures, their functional roles remain unclear. Here, we detected the VFRs in the hippocampal network and tracked their roles during status epilepticus (SE) in rats with pilocarpine-induced temporal lobe epilepsy (TLE). All regions in the hippocampal network exhibited VFRs in the baseline, preictal, ictal and postictal states, with the ictal state containing the most VFRs. Moreover, strong phase-locking couplings existed between VFRs and slow oscillations (1-12 Hz) in the ictal and postictal states for all regions. Further investigation indicated that during VFRs, the build-up of slow oscillations in the ictal state began from the temporal lobe and then spread through the whole hippocampal network via two different pathways, which might be associated with the underlying propagation of epileptiform discharges in the hippocampal network. Overall, we provide a functional description of the emergence of VFRs in the hippocampal network during SE, and we also establish that VFRs may be the physiological representation of the pathological alterations in hippocampal network activity during SE in TLE.

Decision response and feedback in gambling are interrelated. Different decisions lead to different ranges of feedback, which in turn influences subsequent decisions. However, the mechanism underlying the continuous decision-feedback process is still left unveiled. To fulfill this gap, we applied the hidden Markov model (HMM) to the gambling electroencephalogram (EEG) data to characterize the dynamics of this process. Furthermore, we explored the differences between distinct decision responses (i.e. choose large or small bets) or distinct feedback (i.e. win or loss outcomes) in corresponding phases. We demonstrated that the processing stages in decision-feedback process including strategy adjustment and visual information processing can be characterized by distinct brain networks. Moreover, time-varying networks showed, after decision response, large bet recruited more resources from right frontal and right center cortices while small bet was more related to the activation of the left frontal lobe. Concerning feedback, networks of win feedback showed a strong right frontal and right center pattern, while an information flow originating from the left frontal lobe to the middle frontal lobe was observed in loss feedback. Taken together, these findings shed light on general principles of natural decision-feedback and may contribute to the design of biologically inspired, participant-independent decision-feedback systems.

EEG inverse problem is underdetermined, which poses a long standing challenge in Neuroimaging. The combination of source-imaging and analysis of cortical directional networks enables us to noninvasively explore the underlying neural processes. However, existing EEG source imaging approaches mainly focus on performing the direct inverse operation for source estimation, which will be inevitably influenced by noise and the strategy used to find the inverse solution. Here, we develop a new source imaging technique, Deep Brain Neural Network (DeepBraiNNet), for robust sparse spatiotemporal EEG source estimation. In DeepBraiNNet, considering that Recurrent Neural Network (RNN) are usually "deep" in temporal dimension and thus suitable for time sequence modelling, the RNN with Long Short-Term Memory (LSTM) is utilized to approximate the inverse operation for the lead field matrix instead of performing the direct inverse operation, which avoids the possible effect of the direct inverse operation on the underdetermined lead field matrix prone to be influenced by noise. Simulations on various source patterns and noise conditions confirmed that the proposed approach could actually recover the spatiotemporal sources well, outperforming existing state of-the-art methods. DeepBraiNNet also estimated sparse MI related activation patterns when it was applied to a real Motor Imagery dataset, consistent with other findings based on EEG and fMRI. Based on the spatiotemporal sources estimated from DeepBraiNNet, we constructed MI related cortical neural networks, which clearly exhibited strong contralateral network patterns for the two MI tasks. Consequently, DeepBraiNNet may provide an alternative way different from the conventional approaches for spatiotemporal EEG source imaging.

Status epilepticus (SE) is a common, life-threatening neurological disorder that may lead to permanent brain damage. In rodent models, SE is an acute phase of seizures that could be reproduced by injecting with pilocarpine and then induce chronic temporal lobe epilepsy (TLE) seizures. However, how SE disrupts brain activity, especially communications among brain regions, is still unclear. In this study, we aimed to identify the characteristic abnormalities of network connections among the frontal cortex, hippocampus and thalamus during the SE episodes in a pilocarpine model with functional and effective connectivity measurements. We showed that the coherence connectivity among these regions increased significantly during the SE episodes in almost all frequency bands (except the alpha band) and that the frequency band with enhanced connections was specific to different stages of SE episodes. Moreover, with the effective analysis, we revealed a closed neural circuit of bidirectional effective interactions between the frontal regions and the hippocampus and thalamus in both ictal and post-ictal stages, implying aberrant enhancement of communication across these brain regions during the SE episodes. Furthermore, an effective connection from the hippocampus to the thalamus was detected in the delta band during the pre-ictal stage, which shifted in an inverse direction during the ictal stage in the theta band and in the theta, alpha, beta and low-gamma bands during the post-ictal stage. This specificity of the effective connection between the hippocampus and thalamus illustrated that the hippocampal structure is critical for the initiation of SE discharges, while the thalamus is important for the propagation of SE discharges. Overall, our results demonstrated enhanced interaction among the frontal cortex, hippocampus and thalamus during the SE episodes and suggested the modes of information flow across these structures for the initiation and propagation of SE discharges. These findings may reveal an underlying mechanism of aberrant network communication during pilocarpine-induced SE discharges and deepen our knowledge of TLE seizures.

The dynamic profile of brain function has received much attention in recent years and is also a focus in the study of epilepsy. The present study aims to integrate the dynamics of temporal and spatial characteristics to provide comprehensive and novel understanding of epileptic dynamics. Resting state fMRI data were collected from eighty-three patients with idiopathic generalized epilepsy (IGE) and 87 healthy controls (HC). Specifically, we explored the temporal and spatial variation of functional connectivity density (tvFCD and svFCD) in the whole brain. Using a sliding-window approach, for a given region, the standard variation of the FCD series was calculated as the tvFCD and the variation of voxel-wise spatial distribution was calculated as the svFCD. We found primary, high-level, and sub-cortical networks demonstrated distinct tvFCD and svFCD patterns in HC. In general, the high-level networks showed the highest variation, the subcortical and primary networks showed moderate variation, and the limbic system showed the lowest variation. Relative to HC, the patients with IGE showed weaken temporal and enhanced spatial variation in the default mode network and weaken temporospatial variation in the subcortical network. Besides, enhanced temporospatial variation in sensorimotor and high-level networks was also observed in patients. The hyper-synchronization of specific brain networks was inferred to be associated with the phenomenon responsible for the intrinsic propensity of generation and propagation of epileptic activities. The disrupted dynamic characteristics of sensorimotor and high-level networks might potentially contribute to the driven motion and cognition phenotypes in patients. In all, presently provided evidence from the temporospatial variation of functional interaction shed light on the dynamics underlying neuropathological profiles of epilepsy.

Thalamus interacts with cortical areas, generating oscillations characterized by their rhythm and levels of synchrony. However, little is known of what function the rhythmic dynamic may serve in thalamocortical couplings. This work introduced a general approach to investigate the modulatory contribution of rhythmic scalp network to the thalamo-frontal couplings in juvenile myoclonic epilepsy (JME) and frontal lobe epilepsy (FLE). Here, time-varying rhythmic network was constructed using the adapted directed transfer function between EEG electrodes, and then was applied as a modulator in fMRI-based thalamocortical functional couplings. Furthermore, the relationship between corticocortical connectivity and rhythm-dependent thalamocortical coupling was examined. The results revealed thalamocortical couplings modulated by EEG scalp network have frequency-dependent characteristics. Increased thalamus- sensorimotor network (SMN) and thalamus-default mode network (DMN) couplings in JME were strongly modulated by alpha band. These thalamus-SMN couplings demonstrated enhanced association with SMN-related corticocortical connectivity. In addition, altered theta-dependent and beta-dependent thalamus-frontoparietal network (FPN) couplings were found in FLE. The reduced theta-dependent thalamus-FPN couplings were associated with the decreased FPN-related corticocortical connectivity. This study proposed interactive links between the rhythmic modulation and thalamocortical coupling. The crucial role of SMN and FPN in subcortical-cortical circuit may have implications for intervention in generalized and focal epilepsy.

Neuroimaging studies have suggested the presence of abnormalities in the prefrontal-thalamic-cerebellar circuit in schizophrenia (SCH) and depression (DEP). However, the common and distinct structural and causal connectivity abnormalities in this circuit between the two disorders are still unclear. In the current study, structural and resting-state functional magnetic resonance imaging (fMRI) data were acquired from 20 patients with SCH, 20 depressive patients and 20 healthy controls (HC). Voxel-based morphometry analysis was first used to assess gray matter volume (GMV). Granger causality analysis, seeded at regions with altered GMVs, was subsequently conducted. To discover the differences between the groups, ANCOVA and post hoc tests were performed. Then, the relationships between the structural changes, causal connectivity and clinical variables were investigated. Finally, a leave-one-out resampling method was implemented to test the consistency. Statistical analyses showed the GMV and causal connectivity changes in the prefrontal-thalamic-cerebellar circuit. Compared with HC, both SCH and DEP exhibited decreased GMV in middle frontal gyrus (MFG), and a lower GMV in MFG and medial prefrontal cortex (MPFC) in SCH than DEP. Compared with HC, both patient groups showed increased causal flow from the right cerebellum to the MPFC (common causal connectivity abnormalities). And distinct causal connectivity abnormalities (increased causal connectivity from the left thalamus to the MPFC in SCH than HC and DEP, and increased causal connectivity from the right cerebellum to the left thalamus in DEP than HC and SCH). In addition, the structural deficits in the MPFC and its causal connectivity from the cerebellum were associated with the negative symptom severity in SCH. This study found common/distinct structural deficits and aberrant causal connectivity patterns in the prefrontal-thalamic-cerebellar circuit in SCH and DEP, which may provide a potential direction for understanding the convergent and divergent psychiatric pathological mechanisms between SCH and DEP. Furthermore, concomitant structural and causal connectivity deficits in the MPFC may jointly contribute to the negative symptoms of SCH.

Motor imagery (MI) requires subjects to visualize the requested motor behaviors, which involves a large-scale network that spans multiple brain areas. The corresponding cortical activity reflected on the scalp is characterized by event-related desynchronization (ERD) and then by event-related synchronization (ERS). However, the network mechanisms that account for the dynamic information processing of MI during the ERD and ERS periods remain unknown. Here, we combined ERD/ERS analysis with the dynamic networks in different MI stages (i.e. motor preparation, ERD and ERS) to probe the dynamic processing of MI information. Our results show that specific dynamic network structures correspond to the ERD/ERS evolution patterns. Specifically, ERD mainly shows the contralateral networks, while ERS has the symmetric networks. Moreover, different dynamic network patterns are also revealed between the two types of MIs, in which the left-hand MIs exhibit a relatively less sustained contralateral network, which may be the network mechanism that accounts for the bilateral ERD/ERS observed for the left-hand MIs. Similar to the network topologies, the three MI stages also appear to be characterized by different network properties. The above findings all demonstrate that different MI stages that involve specific brain networks for dynamically processing the MI information.

The purpose of this study was to investigate the functional connectivity (FC) of thalamic subdivisions in patients with juvenile myoclonic epilepsy (JME). Resting state functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) data were acquired from 22 JME and 25 healthy controls. We first divided the thalamus into eight subdivisions by performing independent component analysis on tracking fibers and clustering thalamus-related FC maps. We then analyzed abnormal FC in each subdivision in JME compared with healthy controls, and we investigated their associations with clinical features. Eight thalamic sub-regions identified in the current study showed unbalanced thalamic FC in JME: decreased FC with the superior frontal gyrus and enhanced FC with the supplementary motor area in the posterior thalamus increased thalamic FC with the salience network (SN) and reduced FC with the default mode network (DMN). Abnormalities in thalamo-prefrontocortical networks might be related to the propagation of generalized spikes with frontocentral predominance in JME, and the network connectivity differences with the SN and DMN might be implicated in emotional and cognitive defects in JME. JME was also associated with enhanced FC among thalamic sub-regions and with the basal ganglia and cerebellum, suggesting the regulatory role of subcortical nuclei and the cerebellum on the thalamo-cortical circuit. Additionally, increased FC with the pallidum was positive related with the duration of disease. The present study provides emerging evidence of FC to understand that specific thalamic subdivisions contribute to the abnormalities of thalamic-cortical networks in JME. Moreover, the posterior thalamus could play a crucial role in generalized epileptic activity in JME.

Softmax regression, a generalization of Logistic regression (LR) in the setting of multi-class classification, has been widely used in many machine learning applications. However, the performance of softmax regression is extremely sensitive to the presence of noisy data and outliers. To address this issue, we propose a model of robust softmax regression (RoSR) originated from the self-paced learning (SPL) paradigm for multi-class classification. Concretely, RoSR equipped with the soft weighting scheme is able to evaluate the importance of each data instance. Then, data instances participate in the classification problem according to their weights. In this way, the influence of noisy data and outliers (which are typically with small weights) can be significantly reduced. However, standard SPL may suffer from the imbalanced class influence problem, where some classes may have little influence in the training process if their instances are not sensitive to the loss. To alleviate this problem, we design two novel soft weighting schemes that assign weights and select instances locally for each class. Experimental results demonstrate the effectiveness of the proposed methods.

Self-paced learning (SPL), a recently proposed learning strategy, which progressively adds instances to train from simplicity to complexity, could typically reduce the risk of achieving local optima. SPL selects instances based on their losses among the entire data set in each iteration. This probably causes that the selected instances are highly imbalanced, e.g., very few (even on) instances of some classes are chosen, and further negatively affects the training process. To address this issue, we propose a balanced self-paced learning (BSPL) scenario, which iteratively selects training samples based on their loss values from each class, instead of from the entire data set. From another perspective, learning with marginalized corrupted features is an approach to control overfitting by artificially corrupting the training data. However, feature corruption techniques typically lead to that the classification problem is non-convex and easily traps in local optima. To alleviate this, we propose balanced self-paced learning with feature corruption (BSPL-FC), which considers the instance sampling and feature corruption simultaneously. BSPL-FC first treats the feature corruption as a regularizer and then applies BSPL to solve the regularized classification problem. BSPL-FC inherently has advantages in controlling overfitting and avoiding local optima. Experimental results show the effectiveness of the proposed model.

Benign epilepsy with centrotemporal spikes (BECT) is the most common form of childhood idiopathic focal epilepsy syndrome. We investigated quantitative evidence regarding brain morphology and functional connectivity features to provide insight into the neuroanatomical foundation of this disorder, using high resolution T1-weighted magnetic resonance imaging (MRI) and resting state functional MRI in 21 patients with BECT and in 20 healthy children. The functional connectivity analysis, seeded at the regions with altered gray-matter (GM) volume in voxel-based morphometry (VBM) analysis, was further performed. Then, the observed structural and functional alteration were investigated for their association with the clinical and behavior manifestations. The increased GM volume in the striatum and fronto-temporo-parietal cortex (striato-cortical circuit) was observed in BECT. The decreased connections were found among the motor network and frontostriatal loop, and between the default mode network (DMN) and language regions. Additionally, the GM of striatum was negatively correlated with age at epilepsy onset. The current observations may contribute to the understanding of the altered structural and functional feature of striato-cortical circuit in patients with BECT. The findings also implied alterations of the motor network and DMN, which were associated with the epileptic activity in patients with BECT. This further suggested that the onset of BECT might have enduring structural and functional effects on brain maturation.

The basal ganglia (BG), serving as an intermediate bridge between the cerebral cortex and thalamus, are believed to play crucial roles in controlling absence seizure activities generated by the pathological corticothalamic system. Inspired by recent experiments, here we systematically investigate the contribution of a novel identified GABAergic pallido-cortical pathway, projecting from the globus pallidus externa (GPe) in the BG to the cerebral cortex, to the control of absence seizures. By computational modelling, we find that both increasing the activation of GPe neurons and enhancing the coupling strength of the inhibitory pallido-cortical pathway can suppress the bilaterally synchronous 2-4 Hz spike and wave discharges (SWDs) during absence seizures. Appropriate tuning of several GPe-related pathways may also trigger the SWD suppression, through modulating the activation level of GPe neurons. Furthermore, we show that the previously discovered bidirectional control of absence seizures due to the competition between other two BG output pathways also exists in our established model. Importantly, such bidirectional control is shaped by the coupling strength of this direct GABAergic pallido-cortical pathway. Our work suggests that the novel identified pallido-cortical pathway has a functional role in controlling absence seizures and the presented results might provide testable hypotheses for future experimental studies.

To listen to brain activity as a piece of music, we proposed the scale-free brainwave music (SFBM) technology, which could translate the scalp electroencephalogram (EEG) into music notes according to the power law of both EEG and music. In the current study, this methodology was further extended to a musical ensemble of two channels. First, EEG data from two selected channels are translated into musical instrument digital interface (MIDI) sequences, where the EEG parameters modulate the pitch, duration, and volume of each musical note. The phase synchronization index of the two channels is computed by a Hilbert transform. Then the two MIDI sequences are integrated into a chorus according to the phase synchronization index. The EEG with a high synchronization index is represented by more consonant musical intervals, while the low index is expressed by inconsonant musical intervals. The brain ensemble derived from real EEG segments illustrates differences in harmony and pitch distribution during the eyes-closed and eyes-open states. Furthermore, the scale-free phenomena exist in the brainwave ensemble. Therefore, the scale-free brain ensemble modulated by phase synchronization is a new attempt to express the EEG through an auditory and musical way, and it can be used for EEG monitoring and bio-feedback.

Absence epilepsy is believed to be associated with the abnormal interactions between the cerebral cortex and thalamus. Besides the direct coupling, anatomical evidence indicates that the cerebral cortex and thalamus also communicate indirectly through an important intermediate bridge-basal ganglia. It has been thus postulated that the basal ganglia might play key roles in the modulation of absence seizures, but the relevant biophysical mechanisms are still not completely established. Using a biophysically based model, we demonstrate here that the typical absence seizure activities can be controlled and modulated by the direct GABAergic projections from the substantia nigra pars reticulata (SNr) to either the thalamic reticular nucleus (TRN) or the specific relay nuclei (SRN) of thalamus, through different biophysical mechanisms. Under certain conditions, these two types of seizure control are observed to coexist in the same network. More importantly, due to the competition between the inhibitory SNr-TRN and SNr-SRN pathways, we find that both decreasing and increasing the activation of SNr neurons from the normal level may considerably suppress the generation of spike-and-slow wave discharges in the coexistence region. Overall, these results highlight the bidirectional functional roles of basal ganglia in controlling and modulating absence seizures, and might provide novel insights into the therapeutic treatments of this brain disorder.

Intensity-modulated radiotherapy (IMRT) is being increasingly used for treatment of malignant cancer. Beam angle optimization (BAO) is an important problem in IMRT. In this paper, an emerging population-based meta-heuristic algorithm named ant colony optimization (ACO) is introduced to solve the BAO problem. In the proposed algorithm, a multi-layered graph is designed to map the BAO problem to ACO, and a heuristic function based on the beam's-eye-view dosimetrics (BEVD) score is introduced. In order to verify the feasibility of the presented algorithm, a clinical prostate tumor case is employed, and the preliminary results demonstrate that ACO appears more effcient than genetic algorithm (GA) and can find the optimal beam angles within a clinically acceptable computation time.

Interpretation of the clinical electroencephalographs (EEGs) almost always involves speculation as to the possible locations of the sources inside the brain that are responsible for the observed activity on the scalp. Dipoles are widely used to approximate the sources of electrical activity inside the brain. In this paper, we introduce a novel particle swarm optimization (PSO) algorithm to the EEG dipole source localization problem. A three-concentric-shell model is chosen as our head model, and the dipole number is restricted to 2. The 2 dipoles, each of which has 3 position elements, are combined and represented as a 6-element particle. Initialized by randomly setting the positions and velocities, the particle swarm evolves iteratively. Reported here are simulated cases to demonstrate the feasibility of the proposed PSO-based algorithm. Four groups of dipoles with different physiological meanings are chosen as the tested source models. Simulated cases with 10% noise level are also tested. The results show that PSO is feasible and efficient for the source localization in EEG. Furthermore, compared with the generally accepted genetic algorithm (GA), the PSO algorithm appears to be more accurate and needs less computation time

This paper reports a new delay subspace decomposition (DSD) algorithm. Instead of using the canonical zero-delay correlation matrix, the new DSD algorithm introduces a delay into the correlation matrix of the subspace decomposition to suppress noises in the data. The algorithm is applied to functional magnetic resonance imaging (fMRI) to detect the regions of focal activities in the brain. The efficiency is evaluated by comparing with independent component analysis and principal component analysis method of fMRI.