Revealing brain connectivity: graph embeddings for EEG representation learning and comparative analysis of structural and functional connectivity.

Developmental malformations (dysgenesis) of the corpus callosum lead to neurological conditions with a broad range of clinical presentations. Investigating the altered brain connectivity patterns is crucial to understanding both adaptive and maladaptive neuroplasticity in corpus callosum dysgenesis patients. Here, we acquired structural diffusion-weighted and resting-state functional MRI data from a cohort of 11 corpus callosum dysgenesis patients (five with agenesis and six with hypoplasia) and compared their structural and functional connectivity patterns to healthy subjects selected from the Human Connectome Project. We found that these patients have fewer structural inter- and intra-hemispheric brain connections relative to healthy controls. Interestingly, the patients with callosal agenesis have a scant number of inter-hemispheric connections but manage to maintain the full integrity of functional connectivity between the same cortical regions as the healthy subjects. On the other hand, the hypoplasic group presented abnormal structural and functional connectivity patterns relative to healthy controls while maintaining the same total amount of functional connections. These results demonstrate that acallosal patients can compensate for having fewer structural brain connections and present functional adaptation. However, hypoplasics present atypical structural connections to different brain regions, leading to entirely new and abnormal functional brain connectivity patterns.

Motor imagery (MI) is a cognitive process wherein an individual mentally rehearses a specific movement without physically executing it. Recently, MI-based brain-computer interface (BCI) has attracted widespread attention. However, accurate decoding of MI and understanding of neural mechanisms still face huge challenges. These seriously hinder the clinical application and development of BCI systems based on MI. Thus, it is very necessary to develop new methods to decode MI tasks. In this work, we propose a multi-branch convolutional neural network (MBCNN) with a temporal convolutional network (TCN), an end-to-end deep learning framework to decode multi-class MI tasks. We first used MBCNN to capture the MI electroencephalography signals information on temporal and spectral domains through different convolutional kernels. Then, we introduce TCN to extract more discriminative features. The within-subject cross-session strategy is used to validate the classification performance on the dataset of BCI Competition IV-2a. The results showed that we achieved 75.08% average accuracy for 4-class MI task classification, outperforming several state-of-the-art approaches. The proposed MBCNN-TCN-Net framework successfully captures discriminative features and decodes MI tasks effectively, improving the performance of MI-BCIs. Our findings could provide significant potential for improving the clinical application and development of MI-based BCI systems.

The right and left-hand motor imagery (MI) analysis based on the electroencephalogram (EEG) signal can directly link the central nervous system to a computer or a device. This study aims to identify a set of robust and nonlinear effective brain connectivity features quantified by transfer entropy (TE) to characterize the relationship between brain regions from EEG signals and create a hierarchical feature selection and classification for discrimination of right and lefthand MI tasks. TE is calculated among EEG channels as the distinctive, effective connectivity features. TE is a model-free method that can measure nonlinear effective connectivity and analyze multivariate dependent directed information flow among neural EEG channels. Then four feature subset selection methods namely relief-F, Fisher, Laplacian, and local learningbased clustering (LLCFS) algorithms are used to choose the most significant effective connectivity features and reduce redundant information. Finally, support vector machine (SVM) and linear discriminant analysis (LDA) methods are used for classification. Results show that the best performance in 29 healthy subjects and 60 trials is achieved using the TE method via the Relief-F algorithm as feature selection and support vector machine (SVM) classification with 91.02% accuracy. The TE index and a hierarchical feature selection and classification can be useful for the discrimination of right- and left-hand MI tasks from multichannel EEG signals. Effective connectivity features were extracted from electroencephalogram (EEG) to analyze relationships between regions.Four feature selection methods used to select most significant effective features.Support vector machine (SVM) used for discrimination of right and left hand motor imagery (MI) task. In this study, we investigated brain activity using effective connectivity during MI task based on EEG signals. The motor imagery task can accomplish the same goal as motor execution, since they are both activated by the same brain area. Transfer entropy, coherence, and Granger casualty were employed to extract the features. Differential patterns of activity between the left vs. right MI task showed activity around the motor area rather than other areas. In order to reduce redundant information and select the most significant effective connectivity features, four feature subset selection algorithms are used: Relief-F, Fisher, Laplacian, and learning-based clustering feature selection (LLCFS). Then, support vector machine (SVM) and linear discriminant analysis (LDA) are used to classify left and right hand MI task. Comparison of three different connectivity methods showed that TE index had the highest classification accuracy, and could be useful for the discrimination of right and left hand MI tasks from multichannel EEG signals.

The aim of this paper is to classify both (left and right) hand motor imagery (MI) tasks using event-related (de) synchronization (ERD/ERS) patterns based on the inter trial variance (IV). In this work, publicly available online BCI-2003 competition data set was used which contains MI based Electroencephalography (EEG) data of a single subject. The raw MI based EEG signals were extracted from C3 and C4 channels and band pass filtered to mu (8–12 Hz) frequency band. The mean, inter trial variance and ERD/ERS patterns were calculated in both the channels for both MI tasks of the subject. The patterns were used as feature vector and input to the different machine learning classifiers. The logistic regression and Naive Bayesian classifiers were used for the classification of both MI tasks of the subject. For logistic regression classifier, the accuracy, precision and sensitivity were found to be 92.40%, 92.00% and 92.00%, respectively. Meanwhile, 92.19% accuracy, 92.00% precision and 92.00% sensitivity were obtained for the Naive Bayesian classifier. The logistic regression classifier provides maximum classification accuracy (92.40%) in comparison with the existing methods (87.90–90.20%). Result shows that, the proposed method can quantify ERD/ERS patterns effectively in both the channels and also classifies both MI tasks of the subject successfully.

Motor imagery-based brain-computer interface (MI-BCI) currently represents a new trend in rehabilitation. However, individual differences in the responsive frequency bands and a poor understanding of the communication between the ipsilesional motor areas and other regions limit the use of MI-BCI therapy. Objective: Bimanual training has recently attracted attention as it achieves better outcomes as compared to repetitive one-handed training. This study compared the effects of three MI tasks with different visual feedback. Methods: Fourteen healthy subjects performed single hand motor imagery tasks while watching single static hand (traditional MI), single hand with rotation movement (rmMI), and bimanual coordination with a hand pedal exerciser (bcMI). Functional connectivity is estimated by Transfer Entropy (TE) analysis for brain information flow. Results: Brain connectivity of conducting three MI tasks showed that the bcMI demonstrated increased communications from the parietal to the bilateral prefrontal areas and increased contralateral connections between motor-related zones and spatial processing regions. Discussion/Conclusion: The results revealed bimanual coordination operation events increased spatial information and motor planning under the motor imagery task. And the proposed bimanual coordination MI-BCI (bcMI-BCI) can also achieve the effect of traditional motor imagery tasks and promotes more effective connections with different brain regions to better integrate motor-cortex functions for aiding the development of more effective MI-BCI therapy. Clinical and Translational Impact Statement The proposed bcMI-BCI provides more effective connections with different brain areas and integrates motor-cortex functions to promote motor imagery rehabilitation for patients’ impairment.

Brain functional connectivity (FC) changes and microstructural abnormalities are reported in infants born moderate and late preterm (MLPT). We evaluated the effect of low-grade (grades I, II) intraventricular hemorrhage (IVH) in MLPT babies on brain structural connectivity (SC) and FC. Babies born MLPT between January 2014 and May 2017 underwent brain ultrasound (US) at 72h and 7days after birth, and MRI at around term equivalent. The MRI protocol comprised T1- and T2-weighted sequences, diffusion tensor imaging (DTI), and resting-state functional MRI (fMRI). SC and FC were assessed using graph analysis. Of 350 MLPT neonates, 15 showed low-grade IVH on US at 72h, for which brain MRI was available in 10. These 10 infants, with mean gestational age (GA) 34.0 ± 0.8weeks, comprised the study group, and 10 MLPT infants of mean GA 33.9 ± 1.1weeks, with no abnormalities on brain US and MRI, were control subjects. All study subjects presented modularity, small world topology, and rich club organization for both SC and FC. The patients with low-grade IVH had lower FC rich club coefficient and lower SC betweenness centrality in the left frontoparietal operculum, and lower SC rich club coefficient in the right superior orbitofrontal cortex than the control subjects. Topological and functional properties of mature brain connectivity are present in MLPT infants. IVH in these infants was associated with structural and functional abnormalities in the left frontoparietal operculum and right orbitofrontal cortex, regions related to language and cognition.

Motor imagery (MI) is an important brain-computer interface (BCI) paradigm. The traditional MI paradigm (imagining different limbs) limits the intuitive control of the outer devices, while fine MI paradigm (imagining different joint movements from the same limb) can control the mechanical arm without cognitive disconnection. However, the decoding performance of fine MI limits its application. Electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) are widely used in BCI systems because of their portability and easy operation. In this study, a fine MI paradigm including four classes (hand, wrist, shoulder and rest) was designed, and the data of EEG-fNIRS bimodal brain activity was collected from 12 subjects. Event-related desynchronization (ERD) from EEG signals shows a contralateral dominant phenomenon, and there is difference between the ERD of the four classes. For fNIRS signal in the time dimension, the time periods with significant difference can be observed in the activation patterns of four MI tasks. Spatially, the signal peak based brain topographic map also shows difference of these four MI tasks. The EEG signal and fNIRS signal of these four classes are distinguishable. In this study, a bimodal fusion network is proposed to improve the fine MI tasks decoding performance. The features of these two modalities are extracted separately by two feature extractors based on convolutional neural networks (CNN). The recognition performance was significantly improved by the bimodal method proposed in this study, compared with the performance of the single-modal network. The proposed method outperformed all comparison methods, and achieved a four-class accuracy of 58.96%. This paper demonstrates the feasibility of EEG and fNIRS bimodal BCI systems for fine MI, and shows the effectiveness of the proposed bimodal fusion method. This research is supposed to support fine MI-based BCI systems with theories and techniques.

In recent years, deep-learning-based motor imagery (MI) electroencephalography (EEG) decoding methods have shown great potential in the field of the brain–computer interface (BCI). The existing literature is relatively mature in decoding methods for two classes of MI tasks. However, with the increase in MI task classes, decoding studies for four classes of MI tasks need to be further explored. In addition, it is difficult to obtain large-scale EEG datasets. When the training data are limited, deep-learning-based decoding models are prone to problems such as overfitting and poor robustness. In this study, we design a data augmentation method for MI-EEG. The original EEG is slid along the time axis and reconstructed to expand the size of the dataset. Second, we combine the gated recurrent unit (GRU) and convolutional neural network (CNN) to construct a parallel-structured feature fusion network to decode four classes of MI tasks. The parallel structure can avoid temporal, frequency and spatial features interfering with each other. Experimenting on the well-known four-class MI dataset BCI Competition IV 2a shows a global average classification accuracy of 80.7% and a kappa value of 0.74. The proposed method improves the robustness of deep learning to decode small-scale EEG datasets and alleviates the overfitting phenomenon caused by insufficient data. The method can be applied to BCI systems with a small amount of daily recorded data.

Performance of motor imagery (MI)-based brain computer interfaces (BCIs) highly depends on the extraction of features from different brain neural processes. Commonly designed BCIs use band power changes in single channel electroencephalograms (EEGs) to discriminate different MI tasks. In this paper, we studied the information about interactions of spatially separated brain areas by considering the relationships between brain source signals through functional connectivity. We investigated dynamic time-frequency connectivity patterns during four motor imagery movements including left hand, right hand, both feet and tongue by means of Coherence measure estimated from multivariate adaptive autoregressive model coefficients. To tackle the volume conduction problem, sensor space signals were transformed into source space using independent component analysis technique. We showed distinct time-varying Coherence for different motor imageries. For tongue MI movement, bilateral connectivity was observed in hand areas of both hemispheres. The similar connectivity was detected for feet MI task, but with more focus on right hemisphere. During right and left hand MI, bilateral and contralateral brain connectivities were observed respectively. Results provided valuable information for possible classification of MI tasks by considering brain source functional connectivity patterns for BCIs.

To increase the performance Motor imagery (MI) EEG classification, a multi-scale window length algorithm was proposed in this paper. Under this algorithm, eight different time windows with the length of 600–2000 ms were selected to classify the MI data. The confidence level of the classification results from the eight windows were calculated simultaneously, to determine the weights for the outputs of each pre-classifier. The final classification results of MI tasks were generated by the weighted results from the eight windows. Using the proposed algorithm, the response time of MI classification was decreased by 11.4%, comparing to the single window method. On the same level of response time, the classification accuracy was increased by 2.8% by the multi-scale window algorithm. These results show that the proposed algorithm can speed up the detection of MI tasks with reliable classification accuracy.

BackgroundOne of the most promising applications for electroencephalogram (EEG)-based brain computer interface is for stroke rehabilitation. Implemented as a standalone motor imagery (MI) training system or as part of a rehabilitation robotic system, many studies have shown benefits of using them to restore motor control in stroke patients. Hand movements have widely been chosen as MI tasks. Although potentially more challenging to analyze, wrist and forearm movement such as wrist flexion/extension and forearm pronation/supination should also be considered for MI tasks, because these movements are part of the main exercises given to patients in conventional stroke rehabilitation. This paper will evaluate the effectiveness of such movements for MI tasks.MethodsThree hand and wrist movement tasks which were hand opening/closing, wrist flexion/extension and forearm pronation/supination were chosen as motor imagery tasks for both hands. Eleven subjects participated in the experiment. All of them completed hand opening/closing task session. Ten subjects completed two MI task sessions which were hand opening/closing and wrist flexion/extension. Five subjects completed all three MI tasks sessions. Each MI task comprised 8 sessions spanning a 4 weeks period. For classification, feature extraction based on common spatial pattern (CSP) algorithm was used. Two types were implemented, one with conventional CSP (termed WB) and one with an increase number of features achieved by filtering EEG data into five bands (termed FB). Classification was done by linear discriminant analysis (LDA) and support vector machine (SVM).ResultsEight-fold cross validation was applied on EEG data. LDA and SVM gave comparable classification accuracy. FB achieved significantly higher classification accuracy compared to WB. The accuracy of classifying wrist flexion/extension task were higher than that of classifying hand opening/closing task in all subjects. Classifying forearm pronation/supination task achieved higher accuracy than classifying hand opening/closing task in most subjects but achieved lower accuracy than classifying wrist flexion/extension task in all subjects. Significant improvements of classification accuracy were found in nine subjects when considering individual sessions of experiments of all MI tasks. The results of classifying hand opening/closing task and wrist flexion/extension task were comparable to the results of classifying hand opening/closing task and forearm pronation/supination task. Classification accuracy of wrist flexion/extension task and forearm pronation/supination task was lower than those of hand movement tasks and wrist movement tasks.ConclusionHigh classification accuracy of the three MI tasks support the possibility of using EEG-based stroke rehabilitation system with these movements. Either LDA or SVM can equally be chosen as a classifier since the difference of their accuracies is not statistically significant. Significantly higher classification accuracy made FB more suitable for classifying MI task compared to WB. More training sessions could potentially lead to better accuracy as evident in most subjects in this experiment.

With the development of deep learning, many motor imagery brain-computer interfaces based on convolutional neural networks (CNNs) show outstanding performances. However, the trial number of EEG in the training set is usually limited, and redundancy extensively exists in multiple channel EEG. Thus, overfitting often appears in CNN based motor imagery recognition model and greatly affects the performances of model. In this paper, channel drop out is proposed to address this problem by data augmentation and ensemble learning. Specifically, one of all EEG channels will be dropped and replaced by the mean signal of all EEG channels. In this way, the trial number in the training set was enlarged by channel drop out. And at the testing stage, all the EEG trials processed by channel drop out were fed to the CNN model and the average output probabilities of them were applied to determine the prediction. The experiments were conducted on two popular CNN models applied in motor imagery BCI and BCI Competition IV datasets 2a to verify the performances of the proposed channel drop out approach. The results show that average improvements provided by channel drop out in two-category or four-category motor imagery classification are 2.83% and 2.65% compared with the original CNN model. So the channel drop out approach significantly improves the performances of motor imagery based BCI.

In this study, we propose two adaptive frameworks based on evolutionary algorithms for improving the accuracy of motor imagery (MI) task classification. In an MI task, each participant has a different reaction time delay. In this study, the reaction delay was used as a parameter. Specifically, the time point after the reaction delay was defined as the starting point (SP) when MI began. Various classification and optimization algorithms were employed. The methods include the use of a fixed and dynamic SP for each individual, a genetic algorithm, particle swarm optimization (PSO), spatial filter tuning, the use of common spatial patterns, linear discriminant analysis, bagged decision trees, convolutional neural network (CNN), and long short‐term memory (LSTM). Moreover, a dynamic optimization method for multiband filtering (MBF) method for the subbands of EEG signals was developed that improved accuracy by 5.4%. Combining MBF with bagged decision trees with PSO‐optimized parameters achieved the highest accuracy of 82.3%. Moreover, after optimization by PSO, the CNN‐LSTM method achieved a classification accuracy of 81.4% for 1‐s signals, shorter than those in other studies. The method enables the use of EEG signals to control an exoskeleton for patient rehabilitation.

Neuroimaging studies, such as the Human Connectome Project (HCP), often collect multifaceted data to study the human brain. However, these data are often analyzed in a pairwise fashion, which can hinder our understanding of how different brain-related measures interact. In this study, we analyze the multi-block HCP data using data integration via analysis of subspaces (DIVAS). We integrate structural and functional brain connectivity, substance use, cognition, and genetics in an exhaustive five-block analysis. This gives rise to the important finding that genetics is the single data modality most predictive of brain connectivity, outside of brain connectivity itself. Nearly 14% of the variation in functional connectivity (FC) and roughly 12% of the variation in structural connectivity (SC) is attributed to shared spaces with genetics. Moreover, investigations of shared space loadings provide interpretable associations between particular brain regions and drivers of variability. Novel Jackstraw hypothesis tests are developed for the DIVAS framework to establish statistically significant loadings. For example, in the (FC, SC, and substance use) subspace, these novel hypothesis tests highlight largely negative functional and structural connections suggesting the brain's role in physiological responses to increased substance use. Our findings are validated on genetically relevant subjects not studied in the main analysis.

To explore altered structural and functional connectivity and network organization in cerebral palsy (CP), by clinical CP subtype (unilateral spastic, bilateral spastic, dyskinetic, and ataxic CP). PubMed and Embase databases were systematically searched. Extracted data included clinical characteristics, analyses, outcome measures, and results. Sixty-five studies were included, of which 50 investigated structural connectivity, and 20 investigated functional connectivity using functional magnetic resonance imaging (14 studies) or electroencephalography (six studies). Five of the 50 studies of structural connectivity and one of 14 of functional connectivity investigated whole-brain network organization. Most studies included patients with unilateral spastic CP; none included ataxic CP. Differences in structural and functional connectivity were observed between investigated clinical CP subtypes and typically developing individuals on a wide variety of measures, including efferent, afferent, interhemispheric, and intrahemispheric connections. Directions for future research include extending knowledge in underrepresented CP subtypes and methodologies, evaluating the prognostic potential of specific connectivity and network measures in neonates, and understanding therapeutic effects on brain connectivity.