Deep learning with convolutional neural networks for EEG decoding and visualization

Cross-session SSVEP brainprint recognition using attentive multi-sub-band depth identity embedding learning network

Brainprint recognition technology, regarded as a promising biometric technology, encounters challenges stemming from the time-varied, low signal-to-noise ratio of brain signals, such as electroencephalogram (EEG). Steady-state visual evoked potentials (SSVEP) exhibit high signal-to-noise ratio and frequency locking, making them a promising paradigm for brainprint recognition. Consequently, the extraction of time-invariant identity information from SSVEP EEG signals is essential. In this paper, we propose an Attentive Multi-sub-band Depth Identity Embedding Learning Network for stable cross-session SSVEP brainprint recognition. To address the issue of low recognition accuracy across sessions, we introduce the Sub-band Attentive Frequency mechanism, which integrates the frequency-domain relevant characteristics of the SSVEP paradigm and focuses on exploring depth-frequency identity embedding information. Also, we employ Attentive Statistic Pooling to enhance the stability of frequency domain feature distributions across sessions. Extensive experimentation and validation were conducted on two multi-session SSVEP benchmark datasets. The experimental results show that our approach outperforms other state-of-art models on 2-second samples across sessions and has the potential to serve as a benchmark in multi-subject biometric recognition systems.

EEG-based cross-subject passive music pitch perception using deep learning models

Pitch plays an essential role in music perception and forms the fundamental component of melodic interpretation. However, objectively detecting and decoding brain responses to musical pitch perception across subjects remains to be explored. In this study, we employed electroencephalography (EEG) as an objective measure to obtain the neural responses of musical pitch perception. The EEG signals from 34 subjects under hearing violin sounds at pitches G3 and B6 were collected with an efficient passive Go/No-Go paradigm. The lightweight modified EEGNet model was proposed for EEG-based pitch classification. Specifically, within-subject modeling with the modified EEGNet model was performed to construct individually optimized models. Subsequently, based on the within-subject model pool, a classifier ensemble (CE) method was adopted to construct the cross-subject model. Additionally, we analyzed the optimal time window of brain decoding for pitch perception in the EEG data and discussed the interpretability of these models. The experiment results show that the modified EEGNet model achieved an average classification accuracy of 77% for within-subject modeling, significantly outperforming other compared methods. Meanwhile, the proposed CE method achieved an average accuracy of 74% for cross-subject modeling, significantly exceeding the chance-level accuracy of 50%. Furthermore, we found that the optimal EEG data window for the pitch perception lies 0.4 to 0.9 s onset. These promising results demonstrate that the proposed methods can be effectively used in the objective assessment of pitch perception and have generalization ability in cross-subject modeling.

EEG microstate biomarkers for schizophrenia: a novel approach using deep neural networks

Schizophrenia remains a challenging neuropsychiatric disorder with complex diagnostic processes. Current clinical approaches often rely on subjective assessments, highlighting the critical need for objective, quantitative diagnostic methods. This study aimed to develop a robust classification approach for schizophrenia using EEG microstate analysis and advanced machine learning techniques. We analyzed EEG signals from 14 healthy individuals and 14 patients with schizophrenia during a 15-min resting-state session across 19 EEG channels. A data augmentation strategy expanded the dataset to 56 subjects in each group. The signals were preprocessed and segmented into five frequency bands (delta, theta, alpha, beta, gamma) and five microstates (A, B, C, D, E) using k-means clustering. Five key features were extracted from each microstate: duration, occurrence, standard deviation, coverage, and frequency. A Deep Neural Network (DNN) model, along with other machine learning classifiers, was developed to classify the data. A comprehensive fivefold cross-validation approach evaluated model performance across various EEG channels, frequency bands, and feature combinations. Significant alterations in microstate transition probabilities were observed, particularly in higher frequency bands. The gamma band showed the most pronounced differences, with a notable disruption in D → A transitions (absolute difference = 0.100). The Random Forest classifier achieved the highest accuracy of 99.94% ± 0.12%, utilizing theta band features from the F8 frontal channel. The deep neural network model demonstrated robust performance with 98.31% ± 0.68% accuracy, primarily in the occipital region. Feature size 2 consistently provided optimal classification across most models. Our study introduces a novel, high-precision EEG microstate analysis approach for schizophrenia diagnosis, offering an objective diagnostic tool with potential applications in neuropsychiatric disorders. The findings reveal critical insights into neural dynamics associated with schizophrenia, demonstrating the potential for transforming clinical diagnostic practices through advanced machine learning and neurophysiological feature extraction.

Cross-subject mental workload recognition using bi-classifier domain adversarial learning

To deploy Electroencephalogram (EEG) based Mental Workload Recognition (MWR) systems in the real world, it is crucial to develop general models that can be applied across subjects. Previous studies have utilized domain adaptation to mitigate inter-subject discrepancies in EEG data distributions. However, they have focused on reducing global domain discrepancy, while neglecting local workload-categorical domain divergence. This degrades the workload-discriminating ability of subject-invariant features. To deal with this problem, we propose a novel joint category-wise and domain-wise alignment Domain Adaptation (cdaDA) algorithm, using bi-classifier learning and domain discriminative adversarial learning. The bi-classifier learning approach is adopted to address the similarities and differences between categories, helping to align EEG data within the same mental workload categories. Additionally, the domain discriminative adversarial learning technique is adopted to consider global domain information by minimizing global domain discrepancy. By integrating both local category information and global domain information, the cdaDA model performs a coarse-to-fine alignment and achieves promising cross-subject MWR results.

PhyTransformer: A unified framework for learning spatial-temporal representation from physiological signals

As a modal of physiological information, electroencephalogram (EEG), surface electromyography (sEMG), and eye tracking (ET) signals are widely used to decode human intention, promoting the development of human-computer interaction systems. Extensive studies have achieved single-modal signal decoding with substantial structural differences but consuming mass computing resources and development costs. Considering the similarity in data structure and features, this work proposed a unified framework called PhyTransformer that extracts temporal dynamic and complex channel relationships to decode the physiological signals generally. Concretely, PhyTransformer uses a stacked distillation convolution to capture the complementary temporal dynamic representation from local to global. Considering the information fusion between different channels, our method regards the temporal dynamic of each channel as a token and feeds them into the multi-head attention network to model the complex channel relationship. Subsequently, to measure the channel contributions and fuse the representations from different convolution kernels, PhyTransformer adopts a depth-wise and a separable-wise convolution to extract the final spatial-temporal representation. The proposed method has been evaluated on six publicly benchmarked datasets for physiological signal classification, namely THU and GIST for EEG, Ninapro DB 1 and 6 for sEMG, and GazeCom and HMR for ET. Experiment results illustrate that the proposed method PhyTransformer has the ability to learn robust spatial-temporal representations from multiple modal physiological signals. The code is available at https://github.com/Tammie-Li/PhyTransformer.

Enhancing motor imagery EEG classification with a Riemannian geometry-based spatial filtering (RSF) method

Motor imagery (MI) refers to the mental simulation of movements without physical execution, and it can be captured using electroencephalography (EEG). This area has garnered significant research interest due to its substantial potential in brain-computer interface (BCI) applications, especially for individuals with physical disabilities. However, accurate classification of MI EEG signals remains a major challenge due to their non-stationary nature, low signal-to-noise ratio, and sensitivity to both external and physiological noise. Traditional classification methods, such as common spatial pattern (CSP), often assume that the data is stationary and Gaussian, which limits their applicability in real-world scenarios where these assumptions do not hold. These challenges highlight the need for more robust methods to improve classification accuracy in MI-BCI systems. To address these issues, this study introduces a Riemannian geometry-based spatial filtering (RSF) method that projects EEG signals into a lower-dimensional subspace, maximizing the Riemannian distance between covariance matrices from different classes. By leveraging the inherent geometric properties of EEG data, RSF enhances the discriminative power of the features while maintaining robustness against noise. The performance of RSF was evaluated in combination with ten commonly used MI decoding algorithms, including CSP with linear discriminant analysis (CSP-LDA), Filter Bank CSP (FBCSP), Minimum Distance to Riemannian Mean (MDM), Tangent Space Mapping (TSM), EEGNet, ShallowConvNet (sCNN), DeepConvNet (dCNN), FBCNet, Graph-CSPNet, and LMDA-Net, using six publicly available MI-BCI datasets. The results demonstrate that RSF significantly improves classification accuracy and reduces computational time, particularly for deep learning models with high computational complexity. These findings underscore the potential of RSF as an effective spatial filtering approach for MI EEG classification, providing new insights and opportunities for the development of robust MI-BCI systems. The code for this research is available at https://github.com/PLC-TJU/RSF.

Data alignment based adversarial defense benchmark for EEG-based BCIs

Machine learning has been extensively applied to signal decoding in electroencephalogram (EEG)-based brain-computer interfaces (BCIs). While most studies have focused on enhancing the accuracy of EEG-based BCIs, more attention should be given to their security. Recent findings reveal that EEG-based BCIs are vulnerable to adversarial attacks. To address this, we present the first adversarial defense benchmark based on data alignment, aiming to enhance both the accuracy and robustness of EEG-based BCIs. This study evaluates nine adversarial defense approaches (including five defense strategies) across five EEG datasets (covering three paradigms), three neural networks, and four experimental scenarios. Our results show for the first time that integrating data augmentation, data alignment, and robust training can further improve both the accuracy and robustness of BCIs compared to using only one or two of them. Furthermore, we provide insights into the characteristics of various adversarial defense approaches based on EEG data alignment, offering valuable guidance for developing more accurate and secure EEG-based BCIs.

A subject transfer neural network fuses Generator and Euclidean alignment for EEG-based motor imagery classification

BACKGROUND

Brain-computer interface (BCI) facilitates the connection between human brain and computer, enabling individuals to control external devices indirectly through cognitive processes. Although it has great development prospects, the significant difference in EEG signals among individuals hinders users from further utilizing the BCI system.

NEW METHOD

Addressing this difference and improving BCI classification accuracy remain key challenges. In this paper, we propose a transfer learning model based on deep learning to transfer the data distribution from the source domain to the target domain, named a subject transfer neural network combining the Generator with Euclidean alignment (ST-GENN). It consists of three parts: 1) Align the original EEG signals in the Euclidean space; 2) Send the aligned data to the Generator to obtain the transferred features; 3) Utilize the Convolution-attention-temporal (CAT) classifier to classify the transferred features.

RESULTS

The model is validated on BCI competition IV 2a, BCI competition IV 2b and SHU datasets to evaluate its classification performance, and the results are 82.85 %, 86.28 % and 67.2 % for the three datasets, respectively.

COMPARISON WITH EXISTING METHODS

The results have been shown to be robust to subject variability, with the average accuracy of the proposed method outperforming baseline algorithms by ranging from 2.03 % to 15.43 % on the 2a dataset, from 0.86 % to 10.16 % on the 2b dataset and from 3.3 % to 17.9 % on the SHU dataset.

CONCLUSIONS FOR RESEARCH ARTICLES

The advantage of our model lies in its ability to effectively transfer the experience and knowledge of the source domain data to the target domain, thus bridging the gap between them. Our method can improve the practicability of MI-BCI systems.

A novel STA-EEGNet combined with channel selection for classification of EEG evoked in 2D and 3D virtual reality

Virtual reality (VR), particularly through 3D presentations, significantly boosts user engagement and task efficiency in fields such as gaming, education, and healthcare, offering more immersive and interactive experiences than traditional 2D formats. This study investigates EEG classification in response to 2D and 3D VR stimuli to deepen our understanding of the neural mechanisms driving VR interactions, with implications for brain-computer interfaces (BCIs). We introduce STA-EEGNet, an innovative model that enhances EEGNet by incorporating spatial-temporal attention (STA), improving EEG signal classification from VR environments. A one-way analysis of variance (ANOVA) was utilized to optimize channel selection, enhancing model accuracy. Comparative experiments showed that STA-EEGNet surpassed traditional EEGNet, achieving a peak accuracy of 99.78 % with channel selection. These findings highlight the benefits of spatial-temporal attention and optimal channel selection in classifying VR-evoked EEG data. This study underscores the importance of integrating spatial-temporal attention with compact convolutional neural networks like EEGNet, not only improving EEG signal classification but also advancing neural decoding and optimizing BCI applications.

A multi-modal dataset of electroencephalography and functional near-infrared spectroscopy recordings for motor imagery of multi-types of joints from unilateral upper limb

As one of the important brain-computer interface (BCI) paradigms, motor imagery (MI) enables the control of external devices via identification of motor intention by decoding the features of Electroencephalography (EEG). Movement imagination of multi-types of joints from the same limb allows the development of more accurate and intuitive BCI systems. In this work, we reported an open dataset including EEG and functional near-infrared spectroscopy (fNIRS) recordings from 18 subjects performing eight MI tasks from four types of joints including hand open/close, wrist flexion/extension, wrist abduction/adduction, elbow pronation/supination, elbow flexion/extension, shoulder pronation/supination, shoulder abduction/adduction, and shoulder flexion/extension, resulting in a total of 5760 trials. The validity of multi-modal data was verified both from the EEG/fNIRS activation patterns and the classification performance. It is expected that this dataset will facilitate the development and innovation of decoding algorithms for MI of multi-types of joints based on multi-modal EEG-fNIRS data.

Enhancing robustness of spatial filters in motor imagery based brain-computer interface via temporal learning

BACKGROUND

In motor imagery-based brain-computer interface (MI-BCI) EEG decoding, spatial filtering play a crucial role in feature extraction. Recent studies have emphasized the importance of temporal filtering for extracting discriminative features in MI tasks. While many efforts have been made to optimize feature extraction externally, stabilizing features from spatial filtering remains underexplored.

NEW METHOD

To address this problem, we propose an approach to improve the robustness of temporal features by minimizing instability in the temporal domain. Specifically, we utilize Jensen-Shannon divergence to quantify temporal instability and integrate decision variables to construct an objective function that minimizes this instability. Our method enhances the stability of variance and mean values in the extracted features, improving the identification of discriminative features and reducing the effects of instability.

RESULTS

The proposed method was applied to spatial filtering models, and tested on two publicly datasets as well as a self-collected dataset. Results demonstrate that the proposed method significantly boosts classification accuracy, confirming its effectiveness in enhancing temporal feature stability.

COMPARISON WITH EXISTING METHODS

We compared our method with spatial filtering methods, and the-state-of-the-art models. The proposed approach achieves the highest accuracy, with 92.43 % on BCI competition III IVa dataset, 84.45 % on BCI competition IV 2a dataset, and 73.18 % on self-collected dataset.

CONCLUSIONS

Enhancing the instability of temporal features contributes to improved MI-BCI performance. This not only improves classification performance but also provides a stable foundation for future advancements. The proposed method shows great potential for EEG decoding.

Decoding cortical responses from visual input using an endovascular brain-computer interface

Objective.Implantable neural interfaces enable recording of high-quality brain signals that can improve our understanding of brain function. This work examined the feasibility of using a minimally invasive endovascular neural interface (ENI) to record interpretable cortical activity from the visual cortex.Approach. A sheep model (n= 5) was used to record and decode visually evoked potentials from the cortex both with an ENI and a subdural electrode grid. Sets of distinct experimental visual stimuli were presented to attempt decoding from the recorded cortical potentials, using perceptual categories of colour, contrast, movement direction orientation, spatial frequency and temporal frequency. Decoding performances are presented as accuracy scores from K-fold cross-validation of a stratified random forest classification model. The study compared the signal quality and decoding performance between the ENI and electrocorticography (ECoG) electrodes.Main results. Recordings from the ENI array resulted in lower decoding performances than the ECoG array, but the classification scores were significantly above chance in the stimuli categories of colour, contrast, direction and temporal frequency. This study is the first report of visually evoked neural activity using a minimally-invasive ENI.Significance. Overall, the results show that implantable macro-electrodes yield sufficient neural signal definition to discern primary visual percepts, using both endo-vascular and intracranial surgical placements.

Predicting artificial neural network representations to learn recognition model for music identification from brain recordings

Recent studies have demonstrated that the representations of artificial neural networks (ANNs) can exhibit notable similarities to cortical representations when subjected to identical auditory sensory inputs. In these studies, the ability to predict cortical representations is probed by regressing from ANN representations to cortical representations. Building upon this concept, our approach reverses the direction of prediction: we utilize ANN representations as a supervisory signal to train recognition models using noisy brain recordings obtained through non-invasive measurements. Specifically, we focus on constructing a recognition model for music identification, where electroencephalography (EEG) brain recordings collected during music listening serve as input. By training an EEG recognition model to predict ANN representations-representations associated with music identification-we observed a significant improvement in classification accuracy. This study introduces a novel approach to developing recognition models for brain recordings in response to external auditory stimuli. It holds promise for advancing brain-computer interfaces (BCI), neural decoding techniques, and our understanding of music cognition. Furthermore, it provides new insights into the relationship between auditory brain activity and ANN representations.

Evaluating machine- and deep learning approaches for artifact detection in infant EEG: classifier performance, certainty, and training size effects

Electroencephalography (EEG) is essential for studying infant brain activity but is highly susceptible to artifacts due to infants' movements and physiological variability. Manual artifact detection is labor-intensive and subjective, underscoring the need for automated methods. This study evaluates the performance of three machine learning classifiers - Random Forest (RF), Support Vector Machine (SVM), and a deep learning (DL) model - in detecting artifacts in infant EEG data without prior feature extraction. EEG data were collected from 294 infants (mean age 8.34 months) as part of the Bremen Initiative to Foster Early Childhood Development (BRISE). After preprocessing and manual annotation by an expert, a total of 66,851 epochs were analyzed, with 45% labeled as artifacts. The classifiers were trained on filtered EEG data without further feature extraction to directly handle the complex and noisy signals characteristic of infant EEG. Results. indicated that both the RF classifier and the DL model achieved high balanced accuracy scores (.873 and .881, respectively), substantially outperforming the SVM (.756). Further analysis showed that increasing classifier certainty improved accuracy but reduced the amount of data classified, offering a trade-off between precision and data coverage. Additionally, the RF classifier outperformed the DL model with smaller training datasets, while the DL model required larger datasets to achieve optimal performance. These findings demonstrate that RF and DL classifiers can effectively automate artifact detection in infant EEG data, reducing preprocessing time and enhancing consistency across studies. Implementing such automated methods could facilitate the inclusion of EEG in large-scale developmental research and improve reproducibility by standardizing preprocessing pipelines.

Translational approach for dementia subtype classification using convolutional neural network based on EEG connectome dynamics

Dementia spectrum disorders, characterized by progressive cognitive decline, pose a significant global health burden. Early screening and diagnosis are essential for timely and accurate treatment, improving patient outcomes and quality of life. This study investigated dynamic features of resting-state electroencephalography (EEG) functional connectivity to identify characteristic patterns of dementia subtypes, such as Alzheimer's disease (AD) and frontotemporal dementia (FD), and to evaluate their potential as biomarkers. We extracted distinctive statistical features, including mean, variance, skewness, and Shannon entropy, from brain connectivity measures, revealing common alterations in dementia, specifically a generalized disruption of Alpha-band connectivity. Distinctive characteristics were found, including generalized Delta-band hyperconnectivity with increased complexity in AD and disrupted phase-based connectivity in Theta, Beta, and Gamma bands for FD. We also employed a convolutional neural network model, enhanced with these dynamic features, to differentiate between dementia subtypes. Our classification models achieved a multiclass classification accuracy of 93.6% across AD, FD, and healthy control groups. Furthermore, the model demonstrated 97.8% and 96.7% accuracy in differentiating AD and FD from healthy controls, respectively, and 97.4% accuracy in classifying AD and FD in pairwise classification. These establish a high-performance deep learning framework utilizing dynamic EEG connectivity patterns as potential biomarkers, offering a promising approach for early screening and diagnosis of dementia spectrum disorders using EEG. Our findings suggest that analyzing brain connectivity dynamics as a network and during cognitive tasks could offer more valuable information for diagnosis, assessing disease severity, and potentially identifying personalized neurological deficits.

Virtual Electroencephalogram Acquisition: A Review on Electroencephalogram Generative Methods

Driven by the remarkable capabilities of machine learning, brain-computer interfaces (BCIs) are carving out an ever-expanding range of applications across a multitude of diverse fields. Notably, electroencephalogram (EEG) signals have risen to prominence as the most prevalently utilized signals within BCIs, owing to their non-invasive essence, exceptional portability, cost-effectiveness, and high temporal resolution. However, despite the significant strides made, the paucity of EEG data has emerged as the main bottleneck, preventing generalization of decoding algorithms. Taking inspiration from the resounding success of generative models in computer vision and natural language processing arenas, the generation of synthetic EEG data from limited recorded samples has recently garnered burgeoning attention. This paper undertakes a comprehensive and thorough review of the techniques and methodologies underpinning the generative models of the general EEG, namely the variational autoencoder (VAE), the generative adversarial network (GAN), and the diffusion model. Special emphasis is placed on their practical utility in augmenting EEG data. The structural designs and performance metrics of the different generative approaches in various application domains have been meticulously dissected and discussed. A comparative analysis of the strengths and weaknesses of each existing model has been carried out, and prospective avenues for future enhancement and refinement have been put forward.

MT-RCAF: A Multi-Task Residual Cross Attention Framework for EEG-based emotion recognition and mood disorder detection

BACKGROUND AND OBJECTIVE

Prolonged abnormal emotions can gradually evolve into mood disorders such as anxiety and depression, making it critical to study the relationship between emotions and mood disorders to explore the causes of mood disorders. Existing research on EEG-based emotion recognition and mood disorder detection typically treats these two tasks separately, missing potential synergies between them. The purpose is to reveal the relationship between emotions and mood disorders and propose a Multi-Task Residual Cross Attention Framework (MT-RCAF) to enhance both classification performances.

METHODS

In MT-RCAF, the Feature Extraction module extracts specific and shared features for the corresponding tasks. The Residual Multi-head Cross Attention (RMCA) module dynamically adjusts attention weights to explicitly capture both shared and task-specific information, enhancing complementarity and feature sharing. The Gated Multi-embedding (GME) module filters out irrelevant features, improving task-specific performance. Finally, the Task Tower Classification module balances losses across tasks to facilitate both emotion recognition and mood disorder detection.

RESULTS

We conducted experiments on the DEAP dataset Black as well as the self-collected Emotion and Mood Disorder Dataset (EMDD) to validate the effectiveness of MT-RCAF. The results show that the framework gains improvement in strongly correlated task groups, with average accuracy increases of 3.22% for emotion recognition and 3.91% for mood disorder detection, and in generally correlated task groups, with average accuracy increases of 2.87% for valence and 3.34% for arousal. The study also reveals that mood disorders (depression or anxiety) increase sensitivity to negative emotions, and intense emotions enhance mood disorder detection.

CONCLUSION

The study validates the relationship between emotions and mood disorders from a deep-learning perspective and finds that interconnected tasks result in more accurate and robust results.

A systematic review of artificial intelligence techniques based on electroencephalography analysis in the diagnosis of epilepsy disorders: A clinical perspective

In recent years, Artificial Intelligence (AI), with a specific emphasis on attention mechanisms instead of conventional Deep Learning (DL) or Machine Learning (ML), has demonstrated significant applicability across diverse medical domains. This paper redirects its focus from general brain mapping techniques to specifically investigate the impact of AI in the field of epilepsy diagnosis, concentrating exclusively on Electroencephalography (EEG) data. While earlier studies have predominantly centered on the automatic identification and prediction of seizures using EEG records, an emerging body of research delves into the potential of AI techniques to enhance the analysis of EEG data. This systematic review offers a comprehensive overview, commencing with a concise theoretical exposition on Artificial Neural Networks (ANNs) and attention mechanisms. Subsequent sections explore the applications of AI in EEG analysis for epilepsy, covering aspects such as diagnosis, lateralization, automated lesion detection, presurgical evaluation, and the prediction of postsurgical outcomes. The discussion not only highlights the promising aspects of AI in refining clinical practices but also underscores its potential in tailoring individualized treatments for epilepsy, considering it as a network disorder. The paper concludes by addressing limitations, challenges, and proposing future directions for the application of AI in epilepsy research. While acknowledging the transformative potential of this approach, it emphasizes the necessity for greater multicenter collaboration to amass high-quality data and ensure the open accessibility of developed codes and tools. Moreover, the application of AI models in Computer-Aided Diagnosis (CAD) has exhibited significant promise in enhancing the accuracy and efficiency of epilepsy and seizure diagnosis. This integration of advanced technologies contributes to the development of robust tools for clinical decision-making and underscores the potential for AI-driven solutions in neurological healthcare.

An EEG signal smoothing algorithm using upscale and downscale representation. J Neural Eng 2025;22:036012. [PMID: 40306303 DOI: 10.1088/1741-2552/add297] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Objective.Effective smoothing of electroencephalogram (EEG) signals while maintaining the original signal's features is important in EEG signal analysis and brain-computer interface. This paper proposes a novel EEG signal-smoothing algorithm and its potential application in cognitive conflict (CC) processing.Approach.Instead of being processed in the time domain, the input signal is visualized in increasing line width, the representation frame of which is converted into a binary image. An effective thinning algorithm is employed to obtain a unit-width skeleton as the smoothed signal.Main results.Experimental results on data fitting have verified the effectiveness of the proposed approach on different levels of signal-to-noise (SNR) ratio, especially on high noise levels (SNR⩽5 dB), where our fitting error is only 86.4%-90.4% compared to that of its best counterpart. The potential application of the proposed algorithm in EEG-based CC processing is comprehensively evaluated in a classification and a visual inspection task. The employment of the proposed approach in pre-processing the input data has significantly boosted theF1score of state-of-the-art models by more than 1%. The robustness of our algorithm is also evaluated via a visual inspection task, where specific CC peaks, i.e. the prediction error negativity and error-related positive potential (Pe), can be easily observed at multiple line-width levels, while the insignificant ones are eliminated.Significance.These results demonstrated not only the advance of the proposed approach but also its impact on classification accuracy enhancement.

Behavioural and EEG correlates of forward and backward priming-An exploratory study

During affective priming, perception of an emotional "prime stimulus" influences the reaction time to the subsequent emotional "target stimulus". If prime and target have the same valence (congruent trials), reactions to the target are faster than if prime and target have different valences (incongruent trials). Bem introduced a backward priming paradigm in 2011, where first the target was presented and then the prime after the response. Similar to the classical affective forward priming effects, he found faster reaction times in congruent compared to incongruent trials, and interpreted these results as evidence supporting precognition. In the present exploratory study, while measuring EEG, we combined a forward priming paradigm with a related backward priming paradigm, following Bem's study. We analysed the EEG data on a group level (ERPs) and on an individual level (single participants, applying artificial neural networks). We found significantly faster reaction times for congruent compared to incongruent trials in the forward priming experiment (p = 0.0004) but no statistically significant differences in the backward priming experiment (p = 0.12). We also found significant differences in ERP amplitude in the forward priming congruent vs incongruent conditions (P8 electrode: p = 0.0002). Backward priming results show weaker, shorter, and less significant differences between congruent and incongruent trials, with maxima at electrodes P7, P3, CP5, and CP1. The neural network results were very variable across participants in both the backward and forward priming and on average, the accuracy results were at chance level for both the forward priming as well as the backward priming. Our results replicate behavioural findings and extend the EEG findings for forward priming. We did not replicate Bem's backward priming results. These exploratory EEG results are weak, however they give a good starting point for future studies.

Cross-Subject Motor Imagery Electroencephalogram Decoding with Domain Generalization

Decoding motor imagery (MI) electroencephalogram (EEG) signals in the brain-computer interface (BCI) can assist patients in accelerating motor function recovery. To realize the implementation of plug-and-play functionality for MI-BCI applications, cross-subject models are employed to alleviate time-consuming calibration and avoid additional model training for target subjects by utilizing EEG data from source subjects. However, the diversity in data distribution among subjects limits the model's robustness. In this study, we investigate a cross-subject MI-EEG decoding model with domain generalization based on a deep learning neural network that extracts domain-invariant features from source subjects. Firstly, a knowledge distillation framework is adopted to obtain the internally invariant representations based on spectral features fusion. Then, the correlation alignment approach aligns mutually invariant representations between each pair of sub-source domains. In addition, we use distance regularization on two kinds of invariant features to enhance generalizable information. To assess the effectiveness of our approach, experiments are conducted on the BCI Competition IV 2a and the Korean University dataset. The results demonstrate that the proposed model achieves 8.93% and 4.4% accuracy improvements on two datasets, respectively, compared with current state-of-the-art models, confirming that the proposed approach can effectively extract invariant features from source subjects and generalize to the unseen target distribution, hence paving the way for effective implementation of the plug-and-play functionality in MI-BCI applications.

Classifying Obsessive-Compulsive Disorder from Resting-State EEG using Convolutional Neural Networks: A Pilot Study

Objective: Classifying obsessive-compulsive disorder (OCD) using brain data remains challenging. Resting-state electroencephalography (EEG) offers an affordable and noninvasive approach, but traditional machine learning methods have limited its predictive capability. We explored whether convolutional neural networks (CNNs) applied to minimally processed EEG time-frequency representations could offer a solution, effectively distinguishing individuals with OCD from healthy controls. Method: We collected resting-state EEG data from 20 unmedicated participants (10 OCD, 10 healthy controls). Clean, 4-second EEG segments were transformed into time-frequency representations using Morlet wavelets. In a two-step evaluation, we first used a 2D CNN classifier using leave-one-subject-out cross-validation and compared it to a traditional support vector machine (SVM) trained on spectral band power features. Second, using multimodal fusion, we examined whether adding clinical and demographic information improved classification. Results: The CNN achieved strong classification accuracy (82.0%, AUC: 0.86), significantly outperforming the chance-level SVM baseline (49.0%, AUC: 0.45). Most clinical variables did not improve performance beyond the EEG data alone (subject-level accuracy: 80.0%). However, incorporating education level boosted performance notably (accuracy: 85.0%, AUC: 0.89). Conclusion: CNNs applied to resting-state EEG show promise for diagnosing OCD, outperforming traditional machine learning methods. Despite sample size limitations, these findings highlight deep learning's potential in psychiatric applications. Education level emerged as a potentially complementary feature, warranting further investigation in larger, more diverse samples.

Universal semantic feature extraction from EEG signals: a task-independent framework

Objective.Extracting universal, task-independent semantic features from electroencephalography (EEG) signals remains an open challenge. Traditional approaches are often task-specific, limiting their generalization across different EEG paradigms. This study aims to develop a robust, unsupervised framework for learning high-level, task-independent neural representations.Approach.We propose a novel framework integrating convolutional neural networks, AutoEncoders, and Transformers to extract both low-level spatiotemporal patterns and high-level semantic features from EEG signals. The model is trained in an unsupervised manner to ensure adaptability across diverse EEG paradigms, including motor imagery (MI), steady-state visually evoked potentials (SSVEPs), and event-related potentials (ERPs, specifically P300). Extensive analyses, including clustering, correlation, and ablation studies, are conducted to validate the quality and interpretability of the extracted features.Main results.Our method achieves state-of-the-art performance, with average classification accuracies of 83.50% and 84.84% on MI datasets (BCICIV_2a and BCICIV_2b), 98.41% and 99.66% on SSVEP datasets (Lee2019-SSVEP and Nakanishi2015), and an average AUC of 91.80% across eight ERP datasets. t-distributed stochastic neighbor embedding and clustering analyses reveal that the extracted features exhibit enhanced separability and structure compared to raw EEG data. Correlation studies confirm the framework's ability to balance universal and subject-specific features, while ablation results highlight the near-optimality of the selected model configuration.Significance.This work establishes a universal framework for task-independent semantic feature extraction from EEG signals, bridging the gap between conventional feature engineering and modern deep learning methods. By providing robust, generalizable representations across diverse EEG paradigms, this approach lays the foundation for advanced brain-computer interface applications, cross-task EEG analysis, and future developments in semantic EEG processing.

Application of machine learning in predicting consumer behavior and precision marketing

with the intensification of market competition and the complexity of consumer behavior, enterprises are faced with the challenge of how to accurately identify potential customers and improve user conversion rate. This paper aims to study the application of machine learning in consumer behavior prediction and precision marketing. Four models, namely support vector machine (SVM), extreme gradient boosting (XGBoost), categorical boosting (CatBoost), and backpropagation artificial neural network (BPANN), are mainly used to predict consumers' purchase intention, and the performance of these models in different scenarios is verified through experiments. The results show that CatBoost and XGBoost have the best prediction results when dealing with complex features and large-scale data, F1 scores are 0.93 and 0.92 respectively, and CatBoost's ROC AUC reaches the highest value of 0.985. while SVM has an advantage in accuracy rate, but slightly underperformance when dealing with large-scale data. Through feature importance analysis, we identify the significant impact of page views, residence time and other features on purchasing behavior. Based on the model prediction results, this paper proposes the specific application of optimization marketing strategies such as recommendation system, dynamic pricing and personalized advertising. Future research could improve the predictive power of the model by introducing more kinds of unstructured data, such as consumer reviews, images, videos, and social media data. In addition, the use of deep learning models, such as Transformers or Self-Attention Mechanisms, can better capture complex patterns in long time series data.

EEGUnity: Open-Source Tool in Facilitating Unified EEG Datasets Toward Large-Scale EEG Model

The increasing number of dispersed EEG dataset publications and the advancement of large-scale Electroencephalogram (EEG) models have increased the demand for practical tools to manage diverse EEG datasets. However, the inherent complexity of EEG data, characterized by variability in content data, metadata, and data formats, poses challenges for integrating multiple datasets and conducting large-scale EEG model research. To tackle the challenges, this paper introduces EEGUnity, an open-source tool that incorporates modules of "EEG Parser", "Correction", "Batch Processing", and "Large Language Model Boost". Leveraging the functionality of such modules, EEGUnity facilitates the efficient management of multiple EEG datasets, such as intelligent data structure inference, data cleaning, and data unification. In addition, the capabilities of EEGUnity ensure high data quality and consistency, providing a reliable foundation for large-scale EEG data research. EEGUnity is evaluated across 25 EEG datasets from different sources, offering several typical batch processing workflows. The results demonstrate the high performance and flexibility of EEGUnity in parsing and data processing. The project code is publicly available at github.com/Baizhige/EEGUnity.

Automatic diagnostics of electroencephalography pathology based on multi-domain feature fusion

Electroencephalography (EEG) serves as a practical auxiliary tool deployed to diagnose diverse brain-related disorders owing to its exceptional temporal resolution, non-invasive characteristics, and cost-effectiveness. In recent years, with the advancement of machine learning, automated EEG pathology diagnostics methods have flourished. However, most existing methods usually neglect the crucial spatial correlations in multi-channel EEG signals and the potential complementary information among different domain features, both of which are keys to improving discrimination performance. In addition, latent redundant and irrelevant features may cause overfitting, increased model complexity, and other issues. In response, we propose a novel feature-based framework designed to improve the diagnostic accuracy of multi-channel EEG pathology. This framework first applies a multi-resolution decomposition technique and a statistical feature extractor to construct a salient time-frequency feature space. Then, spatial distribution information is channel-wise extracted from this space to fuse with time-frequency features, thereby leveraging their complementarity to the fullest extent. Furthermore, to eliminate the redundancy and irrelevancy, a two-step dimension reduction strategy, including a lightweight multi-view time-frequency feature aggregation and a non-parametric statistical significance analysis, is devised to pick out the features with stronger discriminative ability. Comprehensive examinations of the Temple University Hospital Abnormal EEG Corpus V. 2.0.0 demonstrate that our proposal outperforms state-of-the-art methods, highlighting its significant potential in clinically automated EEG abnormality detection.

Unsupervised Domain Adaptation With Synchronized Self-Training for Cross- Domain Motor Imagery Recognition

Robust decoding performance is essential for the practical deployment of brain-computer interface (BCI) systems. Existing EEG decoding models often rely on large amounts of annotated data collected through specific experimental setups, which fail to address the heterogeneity of data distributions across different domains. This limitation hinders BCI systems from effectively managing the complexity and variability of real-world data. To overcome these challenges, we propose Synchronized Self-Training Domain Adaptation (SSTDA) for cross-domain motor imagery classification. Specifically, SSTDA leverages labeled signals from a source domain and applies self-training to unlabeled signals from a target domain, enabling the simultaneous training of a more robust classifier. The raw EEG signals are mapped into a latent space by a feature extractor for discriminative representation learning. A domain-shared latent space is then learned by optimizing the feature extractor with both source and target samples, using an easy-tohard self-training process. We validate the method with extensive experiments on two public motor imagery datasets: Dataset IIa of BCI Competition IV and the High Gamma dataset. In the inter-subject task, our method achieves classification accuracies of 64.43% and 80.40%, respectively. It also outperforms existing methods in the inter-session task. Moreover, we develope a new six-class motor imagery dataset and achieve test accuracies of 77.09% and 80.18% across different datasets. All experimental results demonstrate that our SSTDA outperforms existing algorithms in inter-session, inter-subject, and inter-dataset validation protocols, highlighting its capability to learn discriminative, domain-invariant representations that enhance EEG decoding performance.

Effects of Different Preprocessing Pipelines on Motor Imagery-Based Brain-Computer Interfaces

In recent years, brain-computer interfaces (BCIs) leveraging electroencephalography (EEG) signals for the control of external devices have garnered increasing attention. The information transfer rate of BCI has been significantly improved by a lot of cutting-edge methods. The exploration of effective preprocessing in brain-computer interfaces, particularly in terms of identifying suitable preprocessing methods and determining the optimal sequence for their application, remains an area ripe for further investigation. To address this gap, this study explores a range of preprocessing techniques, including but not limited to independent component analysis, surface Laplacian, bandpass filtering, and baseline correction, examining their potential contributions and synergies in the context of BCI applications. In this extensive research, a variety of preprocessing pipelines were rigorously tested across four EEG data sets, all of which were pertinent to motor imagery-based BCIs. These tests incorporated five EEG machine learning models, working in tandem with the preprocessing methods discussed earlier. The study's results highlighted that baseline correction and bandpass filtering consistently provided the most beneficial preprocessing effects. From the perspective of online deployment, after testing and time complexity analysis, this study recommends baseline correction, bandpass filtering and surface Laplace as more suitable for online implementation. An interesting revelation of the study was the enhanced effectiveness of the surface Laplacian algorithm when used alongside algorithms that focus on spatial information. Using appropriate processing algorithms, we can even achieve results (92.91% and 88.11%) that exceed the SOTA feature extraction methods in some cases. Such findings are instrumental in offering critical insights for the selection of effective preprocessing pipelines in EEG signal decoding. This, in turn, contributes to the advancement and refinement of brain-computer interface technologies.

A Distributed Neural Network Architecture for Dynamic Sensor Selection With Application to Bandwidth-Constrained Body-Sensor Networks

We propose a dynamic sensor selection approach for deep neural networks (DNNs), which is able to derive an optimal sensor subset selection for each specific input sample instead of a fixed selection for the entire dataset. This dynamic selection is jointly learned with the task model in an end-to-end way, using the Gumbel-Softmax trick to allow the discrete decisions to be learned through standard backpropagation. We then show how we can use this dynamic selection to increase the lifetime of a wireless sensor network (WSN) by imposing constraints on how often each node is allowed to transmit. We further improve performance by including a dynamic spatial filter that makes the task-DNN more robust against the fact that it now needs to be able to handle a multitude of possible node subsets. Finally, we explain how the selection of the optimal channels can be distributed across the different nodes in a WSN. We validate this method on a use case in the context of body-sensor networks, where we use real electroencephalography (EEG) sensor data to emulate an EEG sensor network for motor execution decoding. For this use case, we demonstrate that the distributed algorithm -with only a small amount of cooperation between the nodes- achieves a performance close to the upper bound defined by a fully centralized dynamic selection (maximum absolute decrease of 4% in accuracy). Furthermore, we observe that our dynamic sensor selection framework can achieve large reductions in transmission energy with a limited cost to the task accuracy, validating it as a practical tool for increasing the lifetime of body-sensor networks.

GLEAM: A multimodal deep learning framework for chronic lower back pain detection using EEG and sEMG signals

Low Back Pain (LBP) is the most prevalent musculoskeletal condition worldwide and a leading cause of disability, significantly affecting mobility, work productivity, and overall quality of life. Due to its high prevalence and substantial economic burden, LBP presents a critical global public health challenge that demands innovative diagnostic and therapeutic solutions. This study introduces a novel deep-learning approach for diagnosing LBP intensity using electroencephalography (EEG) signals and surface electromyography (sEMG) signals from back muscles. A GAN-Convolution-Transformer-based model, named GLEAM (GAN-ConvoLution-sElf Attention-ETLSTM), is designed to classify LBP intensity into four categories: no LBP, mild LBP, moderate LBP, and intolerable LBP. A denoising GAN is central to the model's functionality, playing a pivotal role in enhancing the quality of EEG and sEMG signals by removing noise, resulting in cleaner and more accurate input data. Various features are extracted from the GAN-denoised EEG and sEMG signals, and the combined features from both EEG and sEMG are used for LBP detection. After the feature extraction, the CNN is employed to capture local temporal patterns within the data, allowing the model to focus on smaller, region-specific trends in the signals. Subsequently, the self-attention module identifies global correlations among these locally extracted features, enhancing the model's ability to recognize broader patterns. The proposed ETLSTM network performs the final classification, which achieves an impressive LBP detection accuracy of 98.95%. This research presents several innovative contributions: (i) the development of a novel denoising GAN for cleaning EEG and sEMG signals, (ii) the design and integration of a new ETLSTM architecture as a classifier within the GLEAM model, and (iii) the introduction of the GLEAM hybrid deep learning framework, which enables robust and reliable LBP intensity assessment.

SMANet: A Model Combining SincNet, Multi-Branch Spatial-Temporal CNN, and Attention Mechanism for Motor Imagery BCI

Building a brain-computer interface (BCI) based on motor imagery (MI) requires accurately decoding MI tasks, which poses a significant challenge due to individual discrepancy among subjects and low signal-to-noise ratio of EEG signals. We propose an end-to-end deep learning model, Sinc-multibranch-attention network (SMANet), which combines a SincNet, a multibranch spatial-temporal convolutional neural network (MBSTCNN), and an attention mechanism for MI-BCI classification. Firstly, Sinc convolution is utilized as a band-pass filter bank for data filtering; Second, pointwise convolution facilitates the effective integration of feature information across different frequency ranges, thereby enhancing the overall feature expression capability; Next, the resulting data are fed into the MBSTCNN to learn a deep feature representation. Thereafter, the outputs of the MBSTCNN are concatenated and then passed through an efficient channel attention (ECA) module to enhance local channel feature extraction and calibrate feature mapping. Ultimately, the feature maps yielded by ECA are classified using a fully connected layer. This model SMANet enhances discriminative features through a multi-objective optimization scheme that integrates cross-entropy loss and central loss. The experimental outcomes reveal that our model attains an average accuracy of 80.21% on the four-class MI dataset (BCI Competition IV 2a), 84.02% on the two-class MI dataset (BCI Competition IV 2b), and 72.70% on the two-class MI dataset (OpenBMI). These results are superior to those of the current state-of-the-art methods. The SMANet is capable to effectively decoding the spatial-spectral-temporal information of EEG signals, thereby enhancing the performance of MI-BCIs.

SensoriMind-Trans Net: EEG and sensorimotor-driven transformer for athlete potential evaluation

Introduction

In recent years, the integration of electroencephalogram (EEG) and somatosensory data in athlete potential evaluation has garnered increasing attention. Traditional research methods mainly rely on processing EEG signals or motion sensor data independently. While these methods can provide a certain level of performance assessment, they often overlook the synergy between brain activity and physical movement, making it difficult to comprehensively capture an athlete's potential. Moreover, most existing approaches employ shallow models, which fail to fully exploit the temporal dependencies and cross-modal interactions within the data, leading to suboptimal accuracy and robustness in evaluation results.

Methods

To address these issues, this paper proposes a Transformer-based model, SensoriMind-Trans Net, which combines EEG signals and somatosensory data. The model leverages a multi-layer Transformer network to capture the temporal dependencies of EEG signals and utilizes a somatosensory data feature extractor and cross-modal attention alignment mechanism to enhance the comprehensive evaluation of athletes' cognitive and motor abilities.

Results

Experiments conducted on four public datasets demonstrate that our model outperforms several existing state-of-the-art (SOTA) models in terms of accuracy, inference time, and computational efficiency.

Discussion

Showcasing its broad applicability in athlete potential evaluation. This study offers a new solution for athlete data analysis and holds significant implications for future multimodal sports performance assessment.

MCL-SWT: Mirror Contrastive Learning with Sliding Window Transformer for Subject-Independent EEG Recognition

Background: In brain-computer interfaces (BCIs), transformer-based models have found extensive application in motor imagery (MI)-based EEG signal recognition. However, for subject-independent EEG recognition, these models face challenges: low sensitivity to spatial dynamics of neural activity and difficulty balancing high temporal resolution features with manageable computational complexity. The overarching objective is to address these critical issues. Methods: We introduce Mirror Contrastive Learning with Sliding Window Transformer (MCL-SWT). Inspired by left/right hand motor imagery inducing event-related desynchronization (ERD) in the contralateral sensorimotor cortex, we develop a mirror contrastive loss function. It segregates feature spaces of EEG signals from contralateral ERD locations while curtailing variability in signals sharing similar ERD locations. The Sliding Window Transformer computes self-attention scores over high temporal resolution features, enabling efficient capture of global temporal dependencies. Results: Evaluated on benchmark datasets for subject-independent MI EEG recognition, MCL-SWT achieves classification accuracies of 66.48% and 75.62%, outperforming State-of-the-Art models by 2.82% and 2.17%, respectively. Ablation studies validate the efficacy of both the mirror contrastive loss and sliding window mechanism. Conclusions: These findings underscore MCL-SWT's potential as a robust, interpretable framework for subject-independent EEG recognition. By addressing existing challenges, MCL-SWT could significantly advance BCI technology development.

SS-EMERGE - self-supervised enhancement for multidimension emotion recognition using GNNs for EEG

Self-supervised learning (SSL) is a potent method for leveraging unlabelled data. Nonetheless, EEG signals, characterised by their low signal-to-noise ratio and high-frequency attributes, often do not surpass fully-supervised techniques in cross-subject tasks such as Emotion Recognition. Therefore, this study introduces a hybrid SSL framework: Self-Supervised Enhancement for Multidimension Emotion Recognition using Graph Neural Networks (SS-EMERGE). This model enhances cross-subject EEG-based emotion recognition by incorporating Causal Convolutions for temporal feature extraction, Graph Attention Transformers (GAT) for spatial modelling, and Spectral Embedding for spectral domain analysis. The approach utilises meiosis-based contrastive learning for pretraining, followed by fine-tuning with minimal labelled data, thereby enriching dataset diversity and specificity. Evaluations on the widely-used Emotion recognition datasets, SEED and SEED-IV, reveal that SS-EMERGE achieves impressive Leave-One-Subject-Out (LOSO) accuracies of 92.35% and 81.51%, respectively. It also proposes a foundation model pre-trained on combined SEED and SEED-IV datasets, demonstrating performance comparable to individual models. These results emphasise the potential of SS-EMERGE in advancing EEG-based emotion recognition with high accuracy and minimal labelled data.

Multi-scale convolutional transformer network for motor imagery brain-computer interface

Brain-computer interface (BCI) systems allow users to communicate with external devices by translating neural signals into real-time commands. Convolutional neural networks (CNNs) have been effectively utilized for decoding motor imagery electroencephalography (MI-EEG) signals in BCIs. However, traditional CNN-based methods face challenges such as individual variability in EEG signals and the limited receptive fields of CNNs. This study presents the Multi-Scale Convolutional Transformer (MSCFormer) model that integrates multiple CNN branches for multi-scale feature extraction and a Transformer module to capture global dependencies, followed by a fully connected layer for classification. The multi-branch multi-scale CNN structure effectively addresses individual variability in EEG signals, enhancing the model's generalization capabilities, while the Transformer encoder strengthens global feature integration and improves decoding performance. Extensive experiments on the BCI IV-2a and IV-2b datasets show that MSCFormer achieves average accuracies of 82.95% (BCI IV-2a) and 88.00% (BCI IV-2b), with kappa values of 0.7726 and 0.7599 in five-fold cross-validation, surpassing several state-of-the-art methods. These results highlight MSCFormer's robustness and accuracy, underscoring its potential in EEG-based BCI applications. The code has been released in https://github.com/snailpt/MSCFormer .

The Riemannian Means Field Classifier for EEG-Based BCI Data

: A substantial amount of research has demonstrated the robustness and accuracy of the Riemannian minimum distance to mean (MDM) classifier for all kinds of EEG-based brain-computer interfaces (BCIs). This classifier is simple, fully deterministic, robust to noise, computationally efficient, and prone to transfer learning. Its training is very simple, requiring just the computation of a geometric mean of a symmetric positive-definite (SPD) matrix per class. We propose an improvement of the MDM involving a number of power means of SPD matrices instead of the sole geometric mean. By the analysis of 20 public databases, 10 for the motor-imagery BCI paradigm and 10 for the P300 BCI paradigm, comprising 587 individuals in total, we show that the proposed classifier clearly outperforms the MDM, approaching the state-of-the art in terms of performance while retaining the simplicity and the deterministic behavior. In order to promote reproducible research, our code will be released as open source.

EEG Signal Prediction for Motor Imagery Classification in Brain-Computer Interfaces

Brain-computer interfaces (BCIs) based on motor imagery (MI) generally require EEG signals recorded from a large number of electrodes distributed across the cranial surface to achieve accurate MI classification. Not only does this entail long preparation times and high costs, but it also carries the risk of losing valuable information when an electrode is damaged, further limiting its practical applicability. In this study, a signal prediction-based method is proposed to achieve high accuracy in MI classification using EEG signals recorded from only a small number of electrodes. The signal prediction model was constructed using the elastic net regression technique, allowing for the estimation of EEG signals from 22 complete channels based on just 8 centrally located channels. The predicted EEG signals from the complete channels were used for feature extraction and MI classification. The results obtained indicate a notable efficacy of the proposed prediction method, showing an average performance of 78.16% in classification accuracy. The proposed method demonstrated superior performance compared to the traditional approach that used few-channel EEG and also achieved better results than the traditional method based on full-channel EEG. Although accuracy varies among subjects, from 62.30% to an impressive 95.24%, these data indicate the capability of the method to provide accurate estimates from a reduced set of electrodes. This performance highlights its potential to be implemented in practical MI-based BCI applications, thereby mitigating the time and cost constraints associated with systems that require a high density of electrodes.

LGFormer: integrating local and global representations for EEG decoding

Objective.Electroencephalography (EEG) decoding is challenging because of its temporal variability and low signal-to-noise ratio, which complicate the extraction of meaningful information from signals. Although convolutional neural networks (CNNs) effectively extract local features from EEG signals, they are constrained by restricted receptive fields. In contrast, transformers excel at capturing global dependencies through self-attention mechanisms but often require extensive training data and computational resources, which limits their efficiency on EEG datasets with limited samples.Approach.In this paper, we propose LGFormer, a hybrid network designed to efficiently learn both local and global representations for EEG decoding. LGFormer employs a deep attention module to extract global information from EEG signals, dynamically adjusting the focus of CNNs. Subsequently, LGFormer incorporates a local-enhanced transformer, combining the strengths of CNNs and transformers to achieve multiscale perception from local to global. Despite integrating multiple advanced techniques, LGFormer maintains a lightweight design and training efficiency.Main results.LGFormer achieves state-of-the-art performance within 200 training epochs across four public datasets, including motor imagery, cognitive workload, and error-related negativity decoding tasks. Additionally, we propose a novel spatial and temporal attention visualization method, revealing that LGFormer captures discriminative spatial and temporal features, enhancing model interpretability and providing insights into its decision-making process.Significance.In summary, LGFormer demonstrates superior performance while maintaining high training efficiency across different tasks, highlighting its potential as a versatile and practical model for EEG decoding.

Entropy difference-based EEG channel selection technique for automated detection of ADHD

Attention deficit hyperactivity disorder (ADHD) is one of the common neurodevelopmental disorders in children. This paper presents an automated approach for ADHD detection using the proposed entropy difference (EnD)-based encephalogram (EEG) channel selection approach. In the proposed approach, we selected the most significant EEG channels for the accurate identification of ADHD using an EnD-based channel selection approach. Secondly, a set of features is extracted from the selected channels and fed to a classifier. To verify the effectiveness of the channels selected, we explored three sets of features and classifiers. More specifically, we explored discrete wavelet transform (DWT), empirical mode decomposition (EMD) and symmetrically-weighted local binary pattern (SLBP)-based features. To perform automated classification, we have used k-nearest neighbor (k-NN), Ensemble classifier, and support vectors machine (SVM) classifiers. Our proposed approach yielded the highest accuracy of 99.29% using the public database. In addition, the proposed EnD-based channel selection has consistently provided better classification accuracies than the entropy-based channel selection approach. Also, the developed method has outperformed the existing approaches in automated ADHD detection.

Multi-Scale Pyramid Squeeze Attention Similarity Optimization Classification Neural Network for ERP Detection

Event-related potentials (ERPs) can reveal brain activity elicited by external stimuli. Innovative methods to decode ERPs could enhance the accuracy of brain-computer interface (BCI) technology and promote the understanding of cognitive processes. This paper proposes a novel Multi-Scale Pyramid Squeeze Attention Similarity Optimization Classification Neural Network (MS-PSA-SOC) for ERP Detection. The model integrates a multi-scale architecture, self-attention mechanism, and deep metric learning to achieve a more comprehensive, refined, and discriminative feature representation. The MS module aggregates fine-grained local features and global features with a larger receptive field within a multi-scale architecture, effectively capturing the dynamic characteristics of complex oscillatory activities in the brain at different levels of abstraction. This preserves complementary spatiotemporal representation information. The PSA module continues the multi-scale contextual modeling from the previous module and achieves adaptive recalibration of multi-scale features. By employing effective aggregation and selection mechanisms, it highlights key features while suppressing redundant information. The SOC module jointly optimizes similarity metric loss and classification loss, maintaining the feature space distribution while focusing on sample class labels. This optimization of similarity relationships between samples improves the model's generalization ability and robustness. Results from public and self-collected datasets demonstrate that the command recognition accuracy of the MS-PSA-SOC model is at least 3.1% and 2.8% higher than other advanced algorithms, achieving superior performance. Additionally, the method demonstrates a lower standard deviation across both datasets. This study also validated the network parameters based on Shannon's sampling theorem and EEG "microstates" through relevant experiments.

Decoding SSVEP Via Calibration-Free TFA-Net: A Novel Network Using Time-Frequency Features

Brain-computer interfaces (BCIs) based on steady-state visual evoked potential (SSVEP) signals offer high information transfer rates and non-invasive brain-to-device connectivity, making them highly practical. In recent years, deep learning techniques, particularly convolutional neural network (CNN) architectures, have gained prominence in EEG (e.g., SSVEP) decoding because of their nonlinear modeling capabilities and autonomy from manual feature extraction. However, most studies using CNNs employ temporal signals as the input and cannot directly mine the implicit frequency information, which may cause crucial frequency details to be lost and challenges in decoding. By contrast, the prevailing supervised recognition algorithms rely on a lengthy calibration phase to enhance algorithm performance, which could impede the popularization of SSVEP based BCIs. To address these problems, this study proposes the Time-Frequency Attention Network (TFA-Net), a novel CNN model tailored for SSVEP signal decoding without the calibration phase. Additionally, we introduce the Frequency Attention and Channel Recombination modules to enhance ability of TFA-Net to infer finer frequency-wise attention and extract features efficiently from SSVEP in the time-frequency domain. Classification results on a public dataset demonstrated that the proposed TFA-Net outperforms all the compared models, achieving an accuracy of 79.00% $\pm$ 0.27% and information transfer rate of 138.82 $\pm$ 0.78 bits/min with a 1-s data length. TFA-Net represents a novel approach to SSVEP identification as well as time-frequency signal analysis, offering a calibration-free solution that enhances the generalizability and practicality of SSVEP based BCIs.

EEG-DG: A Multi-Source Domain Generalization Framework for Motor Imagery EEG Classification

Motorimagery EEG classification plays a crucial role in non-invasive Brain-Computer Interface (BCI) research. However, the performance of classification is affected by the non-stationarity and individual variations of EEG signals. Simply pooling EEG data with different statistical distributions to train a classification model can severely degrade the generalization performance. To address this issue, the existing methods primarily focus on domain adaptation, which requires access to the test data during training. This is unrealistic and impractical in many EEG application scenarios. In this paper, we propose a novel multi-source domain generalization framework called EEG-DG, which leverages multiple source domains with different statistical distributions to build generalizable models on unseen target EEG data. We optimize both the marginal and conditional distributions to ensure the stability of the joint distribution across source domains and extend it to a multi-source domain generalization framework to achieve domain-invariant feature representation, thereby alleviating calibration efforts. Systematic experiments conducted on a simulative dataset, BCI competition IV 2a, 2b, and OpenBMI datasets, demonstrate the superiority and competitive performance of our proposed framework over other state-of-the-art methods. Specifically, EEG-DG achieves average classification accuracies of 81.79% and 87.12% on datasets IV-2a and IV-2b, respectively, and 78.37% and 76.94% for inter-session and inter-subject evaluations on dataset OpenBMI, which even outperforms some domain adaptation methods.

From Micro to Meso: A Data-Driven Mesoscopic Region Division Method Based on Functional Connectivity for EEG-Based Driver Fatigue Detection

The integration of EEG signals and deep learning methods is emerging as an effective approach for brain fatigue detection, particularly utilizing Graph Neural Networks(GNNs) that excel in capturing complex electrode relationships. A significant challenge within GNNs is the construction of an effective adjacency matrix that enhances spatial information learning. Concurrently, electrode aggregation in EEG has emerged as a pivotal area of research. However, conventional partitioning methods depend on task-specific prior knowledge, limiting their generalizability across diverse tasks. To Address this issue, we propose a novel mesoscopic region division approach for EEG-based driver fatigue detection, leveraging inherent data characteristics and functional connectivity-based GNN. This method adopts a two-stage approach: initially, micro-electrodes exhibiting similar functional connectivity relationships are grouped as "mesoscopic region"; subsequently, all micro-electrodes in the same group are aggregated into virtual meso-electrodes, and the fatigue state classification is subsequently based on the functional connectivity between them. Applied to a public driver fatigue detection dataset, our approach surpasses existing state-of-the-art methods in performance. Additionally, interpretive analysis provides micro and mesoscopic insights into brain regions and neuronal connections associated with alert and fatigued states.

A Fine-grained Hemispheric Asymmetry Network for accurate and interpretable EEG-based emotion classification

In this work, we propose a Fine-grained Hemispheric Asymmetry Network (FG-HANet), an end-to-end deep learning model that leverages hemispheric asymmetry features within 2-Hz narrow frequency bands for accurate and interpretable emotion classification over raw EEG data. In particular, the FG-HANet extracts features not only from original inputs but also from their mirrored versions, and applies Finite Impulse Response (FIR) filters at a granularity as fine as 2-Hz to acquire fine-grained spectral information. Furthermore, to guarantee sufficient attention to hemispheric asymmetry features, we tailor a three-stage training pipeline for the FG-HANet to further boost its performance. We conduct extensive evaluations on two public datasets, SEED and SEED-IV, and experimental results well demonstrate the superior performance of the proposed FG-HANet, i.e. 97.11% and 85.70% accuracy, respectively, building a new state-of-the-art. Our results also reveal the hemispheric dominance under different emotional states and the hemisphere asymmetry within 2-Hz frequency bands in individuals. These not only align with previous findings in neuroscience but also provide new insights into underlying emotion generation mechanisms.

A multi-layer EEG fusion decoding method with channel selection for multi-brain motor imagery

Traditional motor imagery-based single-brain computer interfaces(BCIs) face inherent limitations, such as unstable signals and low recognition accuracy. In contrast, multi-brain BCIs offer a promising solution by leveraging group electroencephalography (EEG) data. This paper presents a novel multi-layer EEG fusion method with channel selection for motor imagery-based multi-brain BCIs. We utilize mutual information convergent cross-mapping (MCCM) to identify channels that the represent causal relationships between brains; this strategy is combined with multiple linear discriminant analysis (MLDA) for decoding intentions via both data-layer and decision-layer strategies. Our experimental results demonstrate that the proposed method improves the accuracy of multi-brain motor imagery decoding by approximately 10% over that of the traditional methods, with a further 3%-5% accuracy increase due to the effective channel selection mechanism.

Developing Brain-Based Bare-Handed Human-Machine Interaction via On-Skin Input

Developing natural, intuitive, and human-centric input systems for mobile human-machine interaction (HMI) poses significant challenges. Existing gaze or gesture-based interaction systems are often constrained by their dependence on continuous visual engagement, limited interaction surfaces, or cumbersome hardware. To address these challenges, we propose MetaSkin, a novel neurohaptic interface that uniquely integrates neural signals with on-skin interaction for bare-handed, eyes-free interaction by exploiting human's natural proprioceptive capabilities. To support the interface, we developed a deep learning framework that employs multiscale temporal-spectral feature representation and selective feature attention to effectively decode neural signals generated by on-skin touch and motion gestures. In experiments with 12 participants, our method achieved offline accuracies of 81.95% for touch location discrimination, 71.00% for motion type identification, and 46.08% for 10-class touch-motion classification. In pseudo-online settings, accuracies reached 99.43% for touch onset detection, and 80.34% and 67.02% for classification of touch location and motion type, respectively. Neurophysiological analyses revealed distinct neural activation patterns in the sensorimotor cortex, underscoring the efficacy of our multiscale approach in capturing rich temporal and spectral dynamics. Future work will focus on optimizing the system for diverse user populations and dynamic environments, with a long-term goal of advancing human-centered, neuroadaptive interfaces for next-generation HMI systems. This work represents a significant step toward a paradigm shift in design of brain-computer interfaces, bridging sensory and motor paradigms for building more sophisticated systems.

Flexible Patched Brain Transformer model for EEG decoding

Decoding the human brain using non-invasive methods is a significant challenge. This study aims to enhance electroencephalography (EEG) decoding by developing of machine learning methods. Specifically, we propose the novel, attention-based Patched Brain Transformer model to achieve this goal. The model exhibits flexibility regarding the number of EEG channels and recording duration, enabling effective pre-training across diverse datasets. We investigate the effect of data augmentation methods and pre-training on the training process. To gain insights into the training behavior, we incorporate an inspection of the architecture. We compare our model with state-of-the-art models and demonstrate superior performance using only a fraction of the parameters. The results are achieved with supervised pre-training, coupled with time shifts as data augmentation for multi-participant classification on motor imagery datasets.

Feature-aware domain invariant representation learning for EEG motor imagery decoding

Electroencephalography (EEG)-based motor imagery (MI) is extensively utilized in clinical rehabilitation and virtual reality-based movement control. Decoding EEG-based MI signals is challenging because of the inherent spatio-temporal variability of the original signal representation, coupled with a low signal-to-noise ratio (SNR), which impedes the extraction of clean and robust features. To address this issue, we propose a multi-scale spatio-temporal domain-invariant representation learning method, termed MSDI. By decomposing the original signal into spatial and temporal components, the proposed method extracts invariant features at multiple scales from both components. To further constrain the representation to invariant domains, we introduce a feature-aware shift operation that resamples the representation based on its feature statistics and feature measure, thereby projecting the features into a domain-invariant space. We evaluate our proposed method via two publicly available datasets, BNCI2014-001 and BNCI2014-004, demonstrating state-of-the-art performance on both datasets. Furthermore, our method exhibits superior time efficiency and noise resistance.

Recognition of brain activities via graph-based long short-term memory-convolutional neural network

Introduction

Human brain activities are always difficult to recognize due to its diversity and susceptibility to disturbance. With its unique capability of measuring brain activities, magnetoencephalography (MEG), as a high temporal and spatial resolution neuroimaging technique, has been used to identify multi-task brain activities. Accurately and robustly classifying motor imagery (MI) and cognitive imagery (CI) from MEG signals is a significant challenge in the field of brain-computer interface (BCI).

Methods

In this study, a graph-based long short-term memory-convolutional neural network (GLCNet) is proposed to classify the brain activities in MI and CI tasks. It was characterized by implementing three modules of graph convolutional network (GCN), spatial convolution and long short-term memory (LSTM) to effectively extract time-frequency-spatial features simultaneously. For performance evaluation, our method was compared with six benchmark algorithms of FBCSP, FBCNet, EEGNet, DeepConvNets, Shallow ConvNet and MEGNet on two public datasets of MEG-BCI and BCI competition IV dataset 3.

Results

The results demonstrated that the proposed GLCNet outperformed other models with the average accuracies of 78.65% and 65.8% for two classification and four classification on the MEG-BCI dataset, respectively.

Discussion

It was concluded that the GLCNet enhanced the model's adaptability in handling individual variability with robust performance. This would contribute to the exploration of brain activates in neuroscience.

A multi-day and high-quality EEG dataset for motor imagery brain-computer interface

A key challenge in developing a robust electroencephalography (EEG)-based brain-computer interface (BCI) is obtaining reliable classification performance across multiple days. In particular, EEG-based motor imagery (MI) BCI faces large variability and low signal-to-noise ratio. To address these issues, collecting a large and reliable dataset is critical for learning of cross-session and cross-subject patterns while mitigating EEG signals inherent instability. In this study, we obtained a comprehensive MI dataset from the 2019 World Robot Conference Contest-BCI Robot Contest. We collected EEG data from 62 healthy participants across three recording sessions. This experiment includes two paradigms: (1) two-class tasks: left and right hand-grasping, (2) three-class tasks: left and right hand-grasping, and foot-hooking. The dataset comprises raw data, and preprocessed data. For the two-class data, an average classification accuracy of 85.32% was achieved using EEGNet, while the three-class data achieved an accuracy of 76.90% using deepConvNet. Different researchers can reuse the dataset according to their needs. We hope that this dataset will significantly advance MI-BCI research, particularly in addressing cross-session and cross-subject challenges.