MSAttNet: Multi-scale attention convolutional neural network for motor imagery classification.

Objective. Deep transfer learning has been widely used to address the nonstationarity of electroencephalogram (EEG) data during motor imagery (MI) classification. However, previous deep learning approaches suffer from limited classification accuracy because the temporal and spatial features cannot be effectively extracted. Approach. Here, we propose a novel end-to-end deep subject adaptation convolutional neural network (SACNN) to handle the problem of EEG-based MI classification. Our proposed model jointly optimizes three modules, i.e. a feature extractor, a classifier, and a subject adapter. Specifically, the feature extractor simultaneously extracts the temporal and spatial features from the raw EEG data using a parallel multiscale convolution network. In addition, we design a subject adapter to reduce the feature distribution shift between the source and target subjects by using the maximum mean discrepancy. By minimizing the classification loss and the distribution discrepancy, the model is able to extract the temporal-spatial features to the prediction of a new subject. Main results. Extensive experiments are carried out on three EEG-based MI datasets, i.e. brain–computer interface (BCI) competition IV dataset IIb, BCI competition III dataset IVa, and BCI competition IV dataset I, and the average accuracy reaches to 86.42%, 81.71% and 79.35% on the three datasets respectively. Furthermore, the statistical analysis also indicates the significant performance improvement of SACNN. Significance. This paper reveals the importance of the temporal-spatial features on EEG-based MI classification task. Our proposed SACNN model can make fully use of the temporal-spatial information to achieve the purpose.

EEG-based motor imagery classification using convolutional neural networks with local reparameterization trick

Classification algorithm for EEG-based motor imagery using hybrid neural network with spatio-temporal convolution and multi-head attention mechanism.

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.

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.

Motor impairment is a critical health issue that restricts disabled people from living their lives normally and with comfort. Detecting motor imagery (MI) in electroencephalography (EEG) signals can make their lives easier. There has been a lot of work on detecting two or four different MI movements, which include bilateral, contralateral, and unilateral upper limb movements. However, there is little research on the challenging problem of detecting more than four motor imagery tasks and unilateral lower limb movements. As a solution to this problem, a spectral-spatio-temporal multiscale network (SSTMNet) has been introduced to detect six imagery tasks. It first performs a spectral analysis of an EEG trial and attends to the salient brain waves (rhythms) using an attention mechanism. Then, the temporal dependency across the entire EEG trial is worked out using a temporal dependency block, resulting in spectral-spatio-temporal features, which are passed to a multiscale block to learn multiscale spectral-–spatio-temporal features. Finally, these features are deeply analyzed by a sequential block to extract high-level features, which are used to detect an MI task. In addition, to deal with the small dataset problem for each MI task, the researchers introduce a data augmentation technique based on Fourier transform, which generates new EEG trials from EEG signals belonging to the same class in the frequency domain, with the idea that the coefficients of the same frequencies must be fused, ensuring label-preserving trials. SSTMNet is thoroughly evaluated on a public-domain benchmark dataset; it achieves an accuracy of 77.52% and an F1-score of 56.19%. t-SNE plots, confusion matrices, and ROC curves are presented, which show the effectiveness of SSTMNet. Furthermore, when it is trained on augmented data generated by the proposed data augmentation method, it results in a better performance, which validates the effectiveness of the proposed technique. The results indicate that its performance is comparable with the state-of-the-art methods. An analysis of the features learned by the model reveals that the block architectural design aids the model in distinguishing between multi-imagery tasks.

The unilateral upper limb classification from fMRI-weighted EEG signals using convolutional neural network

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.

In this study, we used two machine learning algorithms, namely, linear support vector machine (SVM) and convolutional neural network (CNN), to classify the BCI (Brain Computer interface) competition IV-2a 2-class MI (motor imagery) data set which consists of EEG data from 9 subjects. For each subject, 5 sessions of signals from three electrodes (C3, Cz, and C4) were recorded with sampling rate 250Hz. The training data, which consisted of the first 3 sessions, included 400 trials. The evaluation data, which consisted of the last 2 sessions, included 320 trials. Each trial started with gazing at fix cross on screen for 3 seconds followed by a one-second visual cue pointing either to the left or right to instruct the subject for left or right motor imagery over a period of 4 seconds, and then followed by a short break of at least 1.5 seconds. Features were extracted from the 0.5 to 2.5 second signals after the cue for each trial from C3 and C4. Each EEG trial was band pass filtered into different frequency bands, namely, delta (0.5-3Hz), theta (4-8Hz), alpha (8-12Hz), beta bands (13-30Hz), gamma bands (31-60Hz). Those filtered signals were then used as the input data for training the linear SVM. In addition, we generated a 2 by 500 matrix by down sampling the training data from each trial. There are 5760 such matrices in total generated from all subjects and serve as the input data for training CNN and the trained model was evaluated by another 340 matrices from each subject. Our CNN architecture consisted of 2 convolution layer and 2 fully connect layers, and there was a batch normalization layer before the activated layer and a dropout layer with a probability of 50% after the activated layer. The classification accuracies evaluated by averaged kappa values obtained from linear SVM and CNN are 0.5 and 0.621, respectively, suggesting the deep learning CNN method is superior to the classical linear SVM on the EEG classification.

Objective. Classification of electroencephalography (EEG)-based motor imagery (MI) is a crucial non-invasive application in brain–computer interface (BCI) research. This paper proposes a novel convolutional neural network (CNN) architecture for accurate and robust EEG-based MI classification that outperforms the state-of-the-art methods. Approach. The proposed CNN model, namely EEG-inception, is built on the backbone of the inception-time network, which has showed to be highly efficient and accurate for time-series classification. Also, the proposed network is an end-to-end classification, as it takes the raw EEG signals as the input and does not require complex EEG signal-preprocessing. Furthermore, this paper proposes a novel data augmentation method for EEG signals to enhance the accuracy, at least by 3%, and reduce overfitting with limited BCI datasets. Main results. The proposed model outperforms all state-of-the-art methods by achieving the average accuracy of 88.4% and 88.6% on the 2008 BCI Competition IV 2a (four-classes) and 2b datasets (binary-classes), respectively. Furthermore, it takes less than 0.025 s to test a sample suitable for real-time processing. Moreover, the classification standard deviation for nine different subjects achieves the lowest value of 5.5 for the 2b dataset and 7.1 for the 2a dataset, which validates that the proposed method is highly robust. Significance. From the experiment results, it can be inferred that the EEG-inception network exhibits a strong potential as a subject-independent classifier for EEG-based MI tasks.

In recent years, machine learning algorithms have been developing rapidly, becoming increasingly powerful tools in decoding physiological and neural signals. The aim of this dissertation is to develop computational tools, and especially machine learning techniques, to identify the most effective methods for feature extraction and classification of these signals. This is particularly challenging due to the highly non-linear, non-stationery, and artifact- and noise-prone nature of these signals. Among basic human-control tasks, reaching and grasping are ubiquitous in everyday life. I investigated different linear and non-linear dimensionality reduction techniques for feature extraction and classification of electromyography (EMG) during a reach-grasp-lift task. The results highlighted the advantages of completely automated feature-learning by Laplacian Eigenmaps over manual feature engineering, especially when combined with classification, to achieve high accuracy with few training samples. The ability to decode and reduce the complexity of EMG could enable new practical applications for EMG in basic science and in the clinic. It could also help design humanoid and other robots. Beyond EMG, a key objective of my dissertation was to decode brain activity during the decision-making processes that lead to voluntary action. This was based on electroencephalography (EEG) and holds the promise for improved brain-computer interfaces (BCIs), particularly related to motor imagery (MI). We developed an end-to-end convolutional neural network with attentional mechanism together with different data augmentation techniques on two benchmark MI datasets. I also collected a new dataset, recorded using high-density EEG, and containing both MI and motor execution (ME) tasks. This enabled us to directly compare the decodability of MI and ME, investigate optimal channel configurations, and much more. In particular, this facilitates the analysis and decoding of MI on the fly—online and in in real time. Another potential use of EEG is measuring brain activity of people who are floating in body-temperature water in a sensory-deprivation-tank float pod. We compared lying in the float pod versus lying in bed (a control condition). And we found differences between the two, especially in the gamma band. More research is required to understand what these findings mean for levels of stress in the float pod.

IntroductionIn the field of brain-computer interfaces (BCI), motor imagery (MI) classification is a critically important task, with the primary objective of decoding an individual's MI intentions from electroencephalogram (EEG) signals. However, MI decoding faces significant challenges, primarily due to the inherent complex temporal dependencies of EEG signals.MethodsThis paper proposes a temporal fusion attention network (TFANet), which aims to improve the decoding performance of MI tasks by accurately modeling the temporal dependencies in EEG signals. TFANet introduces a multi-scale temporal self-attention (MSTSA) mechanism that captures temporal variation in EEG signals across different time scales, enabling the model to capture both local and global features. Moreover, the model adaptively adjusts the channel weights through a channel attention module, allowing it to focus on key signals related to motor imagery. This further enhances the utilization of temporal features. Moreover, by integrating the temporal depthwise separable convolution fusion network (TDSCFN) module, TFANet reduces computational burden while enhancing the ability to capture temporal patterns.ResultsThe proposed method achieves a within-subject classification accuracy of 84.92% and 88.41% on the BCIC-IV-2a and BCIC-IV-2b datasets, respectively. Furthermore, using a transfer learning approach on the BCIC-IV-2a dataset, a cross-subject classification accuracy of 77.2% is attained.ConclusionThese results demonstrate that TFANet is an effective approach for decoding MI tasks with complex temporal dependencies.

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.

Recent advances in the Brain-Computer Interface (BCI) systems state that the accurate Motor Imagery (MI) classification using Electroencephalogram (EEG) plays a vital role. In this paper, we propose a novel real-time feature extraction and classification architecture for four class MI using a combination of Kullback-Leibler Regularized Riemannian Mean (KLRRM) and Linear SVM. By using the KL regularization, the robustness of the features extracted to the noise and outliers is improved. The performance of the proposed architecture is analyzed on the four class MI dataset 2a from the BCI Competition IV. The performance analysis shows that the proposed architecture achieves an average classification accuracy of 74.43% and 51.53% for both the good and noisy subjects respectively. Also, the emphasis is laid on understanding the performance of regularization, and the improvement of robustness to the noise and outliers is demonstrated using the noisy subjects.

Training an accurate classifier for EEG-based brain-computer interface (BCI) requires EEG data from a large number of users, whereas protecting their data privacy is a critical consideration. Federated learning (FL) is a promising solution to this challenge. This paper proposes Federated classification with local Batch-specific batch normalization and Sharpness-aware minimization (FedBS) for privacy protection in EEG-based motor imagery (MI) classification. FedBS utilizes local batch-specific batch normalization to reduce data discrepancies among different clients, and sharpness-aware minimization optimizer in local training to improve model generalization. Experiments on three public MI datasets using three popular deep learning models demonstrated that FedBS outperformed six state-of-the-art FL approaches. Remarkably, it also outperformed centralized training, which does not consider privacy protection at all. In summary, FedBS protects user EEG data privacy, enabling multiple BCI users to participate in large-scale machine learning model training, which in turn improves the BCI decoding accuracy.