pith. sign in

arxiv: 2604.18095 · v1 · submitted 2026-04-20 · 💻 cs.AI

DSAINet: An Efficient Dual-Scale Attentive Interaction Network for General EEG Decoding

Zhiyuan Ma , Zeyuan Li , Zihao Qiu , Jinhao Li , Lingqin Meng , Xinche Zhang , Yixuan Liu , Xinke Shen
show 1 more author
Sen Song
This is my paper

Pith reviewed 2026-05-10 04:41 UTC · model grok-4.3

classification 💻 cs.AI
keywords EEG decodingdual-scale networkattention mechanismsubject-independentgeneralizable decodertemporal dynamicsbrain signal processing
0
0 comments X

The pith

A dual-scale attentive network enables one EEG decoder to generalize across tasks and datasets using fixed hyperparameters.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes DSAINet to overcome limited generalizability in EEG decoders for different tasks under subject-independent conditions. It shows that diverse temporal patterns in EEG can be handled by processing signals through parallel fine-scale and coarse-scale convolutional branches, then refining them with attention within and between branches before aggregating for prediction. This design allows the same model to outperform specialized baselines on multiple tasks and datasets without changing its architecture or hyperparameters. If this holds, it suggests that general-purpose EEG decoders are feasible without task-specific engineering.

Core claim

DSAINet constructs shared spatiotemporal token representations from raw EEG signals and models diverse temporal dynamics through parallel convolutional branches at fine and coarse scales. These representations are adaptively refined by intra-branch attention to emphasize salient scale-specific patterns and by inter-branch attention to integrate features across scales, followed by adaptive token aggregation for prediction. Experiments demonstrate consistent outperformance over 13 baselines on five tasks across ten datasets under strict subject-independent evaluation, all with identical architecture hyperparameters.

What carries the argument

The dual-scale attentive interaction network, which uses parallel fine- and coarse-scale convolutional branches combined with intra- and inter-branch attention mechanisms to capture and integrate varied temporal dynamics in EEG signals.

If this is right

  • The same architecture can be deployed across new EEG decoding tasks without redesign.
  • Performance gains are achieved with only about 77,000 trainable parameters.
  • Interpretable insights into neurophysiological patterns become available from the attention mechanisms.
  • Subject-independent decoding improves for applications like brain-computer interfaces.
  • Adaptive integration across scales reduces reliance on task-tailored temporal biases.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This method could generalize to other biosignal processing domains with similar multi-scale temporal structures.
  • Fixed-scale branches might limit performance on tasks with very different frequency content, suggesting need for adaptive scale selection in future work.
  • Combining this with transfer learning could further reduce data requirements for new tasks.
  • The efficiency makes it suitable for real-time deployment on edge devices.

Load-bearing premise

That diverse task-relevant temporal dynamics in EEG signals can be adequately captured and integrated using fixed fine- and coarse-scale convolutional branches with attention, without introducing task-specific designs.

What would settle it

A new EEG task with temporal dynamics outside the captured fine and coarse scales where a task-specific model significantly outperforms DSAINet under the same evaluation protocol.

Figures

Figures reproduced from arXiv: 2604.18095 by Jinhao Li, Lingqin Meng, Sen Song, Xinche Zhang, Xinke Shen, Yixuan Liu, Zeyuan Li, Zhiyuan Ma, Zihao Qiu.

Figure 1
Figure 1. Figure 1: Architectural comparison of representative CNN-Transformer hybrid models for EEG decoding, together with the proposed DSAINet. organize temporal modeling. In many existing designs, multi￾scale temporal feature extraction and the adaptive coordination of scale-specific information are not explicitly distinguished. As a result, cross-scale information is often combined through relatively simple fusion strate… view at source ↗
Figure 2
Figure 2. Figure 2: Overall architecture of DSAINet. The model first converts raw EEG into shared spatiotemporal tokens, then captures fine- and coarse￾scale temporal patterns through two parallel temporal convolution branches, with the resulting branch-wise features further refined by intra-branch attention, integrated by inter-branch attention, and finally aggregated by adaptive token pooling for classification. where DWk (… view at source ↗
Figure 3
Figure 3. Figure 3: Impact of architectural hyperparameters on classification performance on BCIC-IV-2a, Zhou2016, OpenBMI, and EEGMat. The first row illustrates the effect of the number of Temporal Convolution blocks in each branch. The second row shows the effect of attention depth. maintain discriminative temporal structure by telling the model not only what patterns appear, but also where they occur in the token sequence.… view at source ↗
Figure 4
Figure 4. Figure 4: Effect of segment length on classification performance on Mumtaz2017, ADFTD, Rockhill2021, and Shin2018. Stars indicate the segment lengths adopted in the main experiments. modules have clearly separated roles, they do not need to be stacked deeply to handle more complex functions. Thus, the empirical results support a compact one-layer attentive design. D. Effect of Segment Length Because several downstre… view at source ↗
Figure 6
Figure 6. Figure 6: Mean saliency maps averaged across all folds (or across all subjects for BCIC-IV-2a). Nine saliency maps from eight representative datasets are shown. For disorder-related datasets, only subjects from the corresponding patient groups are included, i.e., MDD patients for Mumtaz2017, PD patients for Rockhill2021, and AD/FTD patients sepa￾rately for ADFTD. 1) Saliency Maps Visualization: The saliency maps in … view at source ↗
Figure 7
Figure 7. Figure 7: Learned attention patterns of Intra-Branch Attentive Refinement and Inter-Branch Attentive Interaction on BCIC-IV-2a, ADFTD, and Shin2018. Each row shows two intra-branch attention maps and two inter-branch attention maps, illustrating dataset-dependent yet structured branch refinement and cross-branch feature interaction. C4/CP4, with additional frontal emphasis around FCz, FC2, F7, and F8, which is compa… view at source ↗
read the original abstract

In real-world applications of noninvasive electroencephalography (EEG), specialized decoders often show limited generalizability across diverse tasks under subject-independent settings. One central challenge is that task-relevant EEG signals often follow different temporal organization patterns across tasks, while many existing methods rely on task-tailored architectural designs that introduce task-specific temporal inductive biases. This mismatch makes it difficult to adapt temporal modeling across tasks without changing the model configuration. To address these challenges, we propose DSAINet, an efficient dual-scale attentive interaction network for general EEG decoding. Specifically, DSAINet constructs shared spatiotemporal token representations from raw EEG signals and models diverse temporal dynamics through parallel convolutional branches at fine and coarse scales. The resulting representations are then adaptively refined by intra-branch attention to emphasize salient scale-specific patterns and by inter-branch attention to integrate task-relevant features across scales, followed by adaptive token aggregation to yield a compact representation for prediction. Extensive experiments on five downstream EEG decoding tasks across ten public datasets show that DSAINet consistently outperforms 13 representative baselines under strict subject-independent evaluation. Notably, this performance is achieved using the same architecture hyperparameters across datasets. Moreover, DSAINet achieves a favorable accuracy-efficiency trade-off with only about 77K trainable parameters and provides interpretable neurophysiological insights. The code is publicly available at https://github.com/zy0929/DSAINet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces DSAINet, an efficient dual-scale attentive interaction network for general EEG decoding. It constructs shared spatiotemporal token representations from raw EEG signals and models diverse temporal dynamics via parallel fine- and coarse-scale convolutional branches, refined by intra-branch attention for scale-specific patterns and inter-branch attention for cross-scale integration, followed by adaptive token aggregation. The central claim is that this single architecture with fixed hyperparameters consistently outperforms 13 representative baselines on five downstream EEG decoding tasks across ten public datasets under strict subject-independent evaluation, while achieving a favorable accuracy-efficiency trade-off (~77K parameters) and providing interpretable neurophysiological insights; public code is released.

Significance. If the empirical claims hold under rigorous validation, the work would be significant for advancing generalizable, task-agnostic EEG decoding by demonstrating that diverse temporal dynamics can be captured without task-specific inductive biases or hyperparameter tuning. The public code release, emphasis on fixed hyperparameters across datasets, and low parameter count are clear strengths supporting reproducibility and practical deployment in real-world noninvasive EEG applications.

major comments (2)
  1. [Methods (dual-scale branches and preprocessing)] Methods section (dual-scale convolutional branches and preprocessing): The claim that the same architecture hyperparameters (including fixed kernel sizes, strides, and dilations in the fine- and coarse-scale branches) suffice across all ten datasets is load-bearing for the central generalizability result. However, EEG datasets commonly differ in sampling rates (e.g., 250 Hz vs. 1000 Hz) and trial lengths; without explicit resampling to a common rate or adaptive scaling (details on preprocessing pipelines are not provided), the 'fine' and 'coarse' receptive fields correspond to inconsistent neurophysiological time scales per dataset. This introduces a potential unacknowledged dataset-specific bias that the inter-branch attention may not fully compensate for, risking that performance differences reflect preprocessing alignments rather than the model's design.
  2. [Experiments and Results] Experiments section (results and evaluation): The reported consistent outperformance lacks accompanying details on statistical testing (e.g., p-values from paired tests with multiple-comparison corrections across ten datasets and five tasks), full ablation studies isolating the contributions of intra- and inter-branch attention, and confirmation that the 13 baselines were reimplemented with identical preprocessing, data splits, and subject-independent protocols. These omissions are load-bearing because they prevent verification that gains are robust and not artifacts of implementation differences.
minor comments (2)
  1. [Abstract and Introduction] The abstract and introduction refer to 'five downstream EEG decoding tasks' without enumerating them (e.g., motor imagery, P300, etc.); adding a brief list would improve accessibility.
  2. [Methods (attention and aggregation)] The description of 'adaptive token aggregation' and the attention mechanisms would benefit from explicit equations or pseudocode to clarify the operations and dimensions involved.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We sincerely thank the referee for the constructive and detailed feedback. The comments highlight important aspects of methodological clarity and experimental rigor that we have addressed through revisions to strengthen the manuscript's claims on generalizability.

read point-by-point responses

  1. Referee: [Methods (dual-scale branches and preprocessing)] Methods section (dual-scale convolutional branches and preprocessing): The claim that the same architecture hyperparameters (including fixed kernel sizes, strides, and dilations in the fine- and coarse-scale branches) suffice across all ten datasets is load-bearing for the central generalizability result. However, EEG datasets commonly differ in sampling rates (e.g., 250 Hz vs. 1000 Hz) and trial lengths; without explicit resampling to a common rate or adaptive scaling (details on preprocessing pipelines are not provided), the 'fine' and 'coarse' receptive fields correspond to inconsistent neurophysiological time scales per dataset. This introduces a potential unacknowledged dataset-specific bias that the inter-branch attention may not fully compensate for, risking that performance differences reflect preprocessing alignments rather than

    Authors: We thank the referee for this observation, which correctly identifies a gap in the original submission. The manuscript stated that experiments used raw EEG signals with fixed hyperparameters but did not detail the preprocessing pipeline. In the revised version, we have added an explicit 'Preprocessing' subsection in Methods: all datasets were uniformly resampled to 250 Hz with anti-aliasing filters, and trials were standardized to a fixed 4-second duration via padding/truncation. Kernel sizes were selected to align with standard EEG bands at this rate (fine-scale for higher frequencies, coarse for lower). The inter-branch attention is designed to adaptively fuse scales per input, mitigating residual variations. We also added a sensitivity analysis showing performance stability under minor rate perturbations. These revisions clarify that the reported generalizability stems from the architecture rather than hidden preprocessing alignments. revision: yes

  2. Referee: [Experiments and Results] Experiments section (results and evaluation): The reported consistent outperformance lacks accompanying details on statistical testing (e.g., p-values from paired tests with multiple-comparison corrections across ten datasets and five tasks), full ablation studies isolating the contributions of intra- and inter-branch attention, and confirmation that the 13 baselines were reimplemented with identical preprocessing, data splits, and subject-independent protocols. These omissions are load-bearing because they prevent verification that gains are robust and not artifacts of implementation differences.

    Authors: We agree these details are essential for rigorous validation. The revised manuscript now includes a dedicated 'Statistical Analysis' paragraph reporting paired Wilcoxon signed-rank tests with Bonferroni correction across all 10 datasets and 5 tasks; all key comparisons yield p < 0.05 and are tabulated. We have expanded the ablation studies into a full table isolating intra-branch attention, inter-branch attention, and dual-scale design, with quantitative contributions shown. For baselines, we confirm reimplementation followed original descriptions using identical subject-independent splits and the now-detailed preprocessing pipeline; the public GitHub repository includes all baseline code for direct verification. These additions eliminate ambiguity regarding robustness. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical claims rest on independent cross-dataset tests

full rationale

The paper proposes DSAINet with dual-scale convolutional branches plus intra- and inter-branch attention, then reports empirical outperformance on five tasks across ten datasets under subject-independent splits using fixed hyperparameters. No derivation, equation, or 'prediction' is presented that reduces by construction to fitted inputs, self-citations, or renamed known results. The central claim is a standard ML generalization result evaluated on held-out data; the architecture choices are explicit design decisions, not tautological. This matches the default non-circular case for an empirical architecture paper.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the empirical effectiveness of the proposed dual-scale attentive interaction mechanism, which is introduced as a new design rather than derived from first principles or external benchmarks.

axioms (1)
  • domain assumption EEG signals contain task-relevant information organized at multiple distinct temporal scales that can be separated by parallel convolutional branches.
    Invoked in the description of the dual-scale branches as the motivation for the architecture.

pith-pipeline@v0.9.0 · 5568 in / 1199 out tokens · 32607 ms · 2026-05-10T04:41:46.128054+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 53 canonical work pages

  1. [1]

    Brain computer interfaces, a review,

    L. F. Nicolas-Alonso and J. Gomez-Gil, “Brain computer interfaces, a review,”sensors, vol. 12, no. 2, pp. 1211–1279, 2012

    work page 2012

  2. [2]

    Electroencephalog- raphy,

    A. Biasiucci, B. Franceschiello, and M. M. Murray, “Electroencephalog- raphy,”Current Biology, vol. 29, no. 3, pp. R80–R85, 2019

    work page 2019

  3. [3]

    EEG correlates of task engagement and mental workload in vigilance, learn- ing, and memory tasks,

    C. Berka, D. J. Levendowski, M. N. Lumicao, A. Yau, G. Davis, V . T. Zivkovic, R. E. Olmstead, P. D. Tremoulet, and P. L. Craven, “EEG correlates of task engagement and mental workload in vigilance, learn- ing, and memory tasks,”Aviation, space, and environmental medicine, vol. 78, no. 5, pp. B231–B244, 2007

    work page 2007

  4. [4]

    BENDR: Using trans- formers and a contrastive self-supervised learning task to learn from massive amounts of EEG data,

    D. Kostas, S. Aroca-Ouellette, and F. Rudzicz, “BENDR: Using trans- formers and a contrastive self-supervised learning task to learn from massive amounts of EEG data,”Frontiers in Human Neuroscience, vol. 15, p. 653659, 2021

    work page 2021

  5. [5]

    EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces,

    V . J. Lawhern, A. J. Solon, N. R. Waytowich, S. M. Gordon, C. P. Hung, and B. J. Lance, “EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces,”Journal of neural engineering, vol. 15, no. 5, p. 056013, 2018

    work page 2018

  6. [6]

    LMDA-Net: A lightweight multi-dimensional attention network for general EEG-based brain- computer interfaces and interpretability,

    Z. Miao, M. Zhao, X. Zhang, and D. Ming, “LMDA-Net: A lightweight multi-dimensional attention network for general EEG-based brain- computer interfaces and interpretability,”NeuroImage, vol. 276, p. 120209, 2023

    work page 2023

  7. [7]

    A com- prehensive review of EEG-based brain–computer interface paradigms,

    R. Abiri, S. Borhani, E. W. Sellers, Y . Jiang, and X. Zhao, “A com- prehensive review of EEG-based brain–computer interface paradigms,” Journal of neural engineering, vol. 16, no. 1, p. 011001, 2019

    work page 2019

  8. [8]

    ESTformer: Transformer utilising spatiotemporal dependencies for electroencephalogram super- resolution,

    D. Li, Z. Zeng, Z. Wang, and H. Yang, “ESTformer: Transformer utilising spatiotemporal dependencies for electroencephalogram super- resolution,”Knowledge-Based Systems, vol. 317, p. 113345, 2025

    work page 2025

  9. [9]

    Quantifying Emotional Patterns for EEG-based Emotion Recognition: An Interpretable Study on EEG Individual Differences,

    H. Chen, X. Li, X. Shao, H. He, J. Li, J. Zhu, S. Sun, and B. Hu, “Quantifying Emotional Patterns for EEG-based Emotion Recognition: An Interpretable Study on EEG Individual Differences,”IEEE Transac- tions on Affective Computing, 2025

    work page 2025

  10. [10]

    Intra-and inter-subject variability in EEG- based sensorimotor brain computer interface: a review,

    S. Saha and M. Baumert, “Intra-and inter-subject variability in EEG- based sensorimotor brain computer interface: a review,”Frontiers in computational neuroscience, vol. 13, p. 87, 2020

    work page 2020

  11. [11]

    Discrepancy between inter-and intra-subject variability in EEG-based motor imagery brain-computer interface: Ev- idence from multiple perspectives,

    G. Huang, Z. Zhao, S. Zhang, Z. Hu, J. Fan, M. Fu, J. Chen, Y . Xiao, J. Wang, and G. Dan, “Discrepancy between inter-and intra-subject variability in EEG-based motor imagery brain-computer interface: Ev- idence from multiple perspectives,”Frontiers in neuroscience, vol. 17, p. 1122661, 2023

    work page 2023

  12. [12]

    Integrating biological and machine intelligence: Attention mechanisms in brain–computer interfaces,

    J. Wang, W. Ye, J. He, L. Zhang, G. Huang, Z. Yu, and Z. Liang, “Integrating biological and machine intelligence: Attention mechanisms in brain–computer interfaces,”Information Fusion, vol. 125, p. 103417, 2026

    work page 2026

  13. [13]

    Status of deep learning for EEG-based brain–computer interface appli- cations,

    K. M. Hossain, M. A. Islam, S. Hossain, A. Nijholt, and M. A. R. Ahad, “Status of deep learning for EEG-based brain–computer interface appli- cations,”Frontiers in computational neuroscience, vol. 16, p. 1006763, 2023

    work page 2023

  14. [14]

    ADFCNN: attention-based dual-scale fusion convolutional neural network for motor imagery brain–computer interface,

    W. Tao, Z. Wang, C. M. Wong, Z. Jia, C. Li, X. Chen, C. P. Chen, and F. Wan, “ADFCNN: attention-based dual-scale fusion convolutional neural network for motor imagery brain–computer interface,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 32, pp. 154–165, 2024

    work page 2024

  15. [15]

    A parallel multi-scale time-frequency block convolutional neural network based on channel attention module for motor imagery classification,

    H. Li, H. Chen, Z. Jia, R. Zhang, and F. Yin, “A parallel multi-scale time-frequency block convolutional neural network based on channel attention module for motor imagery classification,”Biomedical Signal Processing and Control, vol. 79, p. 104066, 2023

    work page 2023

  16. [16]

    EEG conformer: Convolutional transformer for EEG decoding and visualization,

    Y . Song, Q. Zheng, B. Liu, and X. Gao, “EEG conformer: Convolutional transformer for EEG decoding and visualization,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 710–719, 2023

    work page 2023

  17. [17]

    TMSA-Net: A novel attention mechanism for improved motor imagery EEG signal processing,

    Q. Zhao and W. Zhu, “TMSA-Net: A novel attention mechanism for improved motor imagery EEG signal processing,”Biomedical Signal Processing and Control, vol. 102, p. 107189, 2025

    work page 2025

  18. [18]

    MSVTNet: Multi-scale vision transformer neural network for EEG-based motor imagery decoding,

    K. Liu, T. Yang, Z. Yu, W. Yi, H. Yu, G. Wang, and W. Wu, “MSVTNet: Multi-scale vision transformer neural network for EEG-based motor imagery decoding,”IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 12, pp. 7126–7137, 2024

    work page 2024

  19. [19]

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

    W. Zhao, B. Zhang, H. Zhou, D. Wei, C. Huang, and Q. Lan, “Multi-scale convolutional transformer network for motor imagery brain- computer interface,”Scientific Reports, vol. 15, no. 1, p. 12935, 2025

    work page 2025

  20. [20]

    DBConformer: Dual-branch convolutional transformer for EEG decoding,

    Z. Wang, H. Wang, T. Jia, X. He, S. Li, and D. Wu, “DBConformer: Dual-branch convolutional transformer for EEG decoding,”IEEE Jour- nal of Biomedical and Health Informatics, 2025. MAet al.: DSAINET: AN EFFICIENT DUAL-SCALE ATTENTIVE INTERACTION NETWORK FOR GENERAL EEG DECODING 13

    work page 2025

  21. [21]

    CT-MIFNet: Convolutional transformer-based multi-view interaction and fusion net- work for EEG decoding,

    Y . Xiong, J. Li, Y . Zhuang, X. Zhao, Y . Xu, and L. Jie, “CT-MIFNet: Convolutional transformer-based multi-view interaction and fusion net- work for EEG decoding,”Biomedical Signal Processing and Control, vol. 112, p. 108421, 2026

    work page 2026

  22. [22]

    EEG-Deformer: A dense convolutional transformer for brain-computer interfaces,

    Y . Ding, Y . Li, H. Sun, R. Liu, C. Tong, C. Liu, X. Zhou, and C. Guan, “EEG-Deformer: A dense convolutional transformer for brain-computer interfaces,”IEEE Journal of Biomedical and Health Informatics, vol. 29, no. 3, pp. 1909–1918, 2025

    work page 1909

  23. [23]

    Deep learning with convolutional neural networks for EEG decoding and visualization,

    R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fiederer, M. Glasstetter, K. Eggensperger, M. Tangermann, F. Hutter, W. Burgard, and T. Ball, “Deep learning with convolutional neural networks for EEG decoding and visualization,”Human brain mapping, vol. 38, no. 11, pp. 5391– 5420, 2017

    work page 2017

  24. [24]

    Learning temporal information for brain-computer interface using convolutional neural networks,

    S. Sakhavi, C. Guan, and S. Yan, “Learning temporal information for brain-computer interface using convolutional neural networks,”IEEE transactions on neural networks and learning systems, vol. 29, no. 11, pp. 5619–5629, 2018

    work page 2018

  25. [25]

    A channel-projection mixed-scale convolutional neural network for motor imagery EEG decoding,

    Y . Li, X.-R. Zhang, B. Zhang, M.-Y . Lei, W.-G. Cui, and Y .-Z. Guo, “A channel-projection mixed-scale convolutional neural network for motor imagery EEG decoding,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 27, no. 6, pp. 1170–1180, 2019

    work page 2019

  26. [26]

    Subject-independent brain–computer interfaces based on deep convolutional neural networks,

    O.-Y . Kwon, M.-H. Lee, C. Guan, and S.-W. Lee, “Subject-independent brain–computer interfaces based on deep convolutional neural networks,” IEEE transactions on neural networks and learning systems, vol. 31, no. 10, pp. 3839–3852, 2020

    work page 2020

  27. [27]

    Motor imagery EEG classification based on a weighted multi-branch structure suitable for multisubject data,

    H. Wang, J. Jiang, J. Q. Gan, and H. Wang, “Motor imagery EEG classification based on a weighted multi-branch structure suitable for multisubject data,”IEEE Transactions on Biomedical Engineering, vol. 70, no. 11, pp. 3040–3051, 2023

    work page 2023

  28. [28]

    MFRC-Net: Multi- scale feature residual convolutional neural network for motor imagery decoding,

    X. Li, Z. Yang, X. Tu, J. Wang, and J. Huang, “MFRC-Net: Multi- scale feature residual convolutional neural network for motor imagery decoding,”IEEE Journal of Biomedical and Health Informatics, vol. 29, no. 1, pp. 224–234, 2025

    work page 2025

  29. [29]

    EEG- based emotion recognition via channel-wise attention and self attention,

    W. Tao, C. Li, R. Song, J. Cheng, Y . Liu, F. Wan, and X. Chen, “EEG- based emotion recognition via channel-wise attention and self attention,” IEEE Transactions on Affective Computing, vol. 14, no. 1, pp. 382–393, 2023

    work page 2023

  30. [30]

    Classification of motor imagery based on multi-scale feature extraction and the channel- temporal attention module,

    R. Wu, J. Jin, I. Daly, X. Wang, and A. Cichocki, “Classification of motor imagery based on multi-scale feature extraction and the channel- temporal attention module,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 3075–3085, 2023

    work page 2023

  31. [31]

    A functional connectivity- based model with a lightweight attention mechanism for depression recognition using EEG signals,

    M. Ying, J. Zhu, X. Li, and B. Hu, “A functional connectivity- based model with a lightweight attention mechanism for depression recognition using EEG signals,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 32, pp. 4240–4248, 2024

    work page 2024

  32. [32]

    Dynamic-attention-based eeg state transition modeling for emotion recognition,

    X. Shen, R. Gan, K. Wang, S. Yang, Q. Zhang, Q. Liu, D. Zhang, and S. Song, “Dynamic-attention-based eeg state transition modeling for emotion recognition,”IEEE Transactions on Affective Computing, vol. 16, no. 4, pp. 3552–3568, 2025

    work page 2025

  33. [33]

    CTNet: a convolu- tional transformer network for EEG-based motor imagery classification,

    W. Zhao, X. Jiang, B. Zhang, S. Xiao, and S. Weng, “CTNet: a convolu- tional transformer network for EEG-based motor imagery classification,” Scientific reports, vol. 14, no. 1, p. 20237, 2024

    work page 2024

  34. [34]

    Multiscale convolu- tional transformer for EEG classification of mental imagery in different modalities,

    H.-J. Ahn, D.-H. Lee, J.-H. Jeong, and S.-W. Lee, “Multiscale convolu- tional transformer for EEG classification of mental imagery in different modalities,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 646–656, 2023

    work page 2023

  35. [35]

    Review of the BCI competition IV,

    M. Tangermann, K.-R. M ¨uller, A. Aertsen, N. Birbaumer, C. Braun, C. Brunner, R. Leeb, C. Mehring, K. J. Miller, G. R. M ¨uller-Putzet al., “Review of the BCI competition IV,”Frontiers in neuroscience, vol. 6, p. 55, 2012

    work page 2012

  36. [36]

    Brain–computer communication: motivation, aim, and impact of exploring a virtual apartment,

    R. Leeb, F. Lee, C. Keinrath, R. Scherer, H. Bischof, and G. Pfurtscheller, “Brain–computer communication: motivation, aim, and impact of exploring a virtual apartment,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 15, no. 4, pp. 473–482, 2007

    work page 2007

  37. [37]

    A fully automated trial selection method for optimization of motor imagery based brain- computer interface,

    B. Zhou, X. Wu, Z. Lv, L. Zhang, and X. Guo, “A fully automated trial selection method for optimization of motor imagery based brain- computer interface,”PloS one, vol. 11, no. 9, p. e0162657, 2016

    work page 2016

  38. [38]

    EEG dataset and OpenBMI toolbox for three BCI paradigms: An investigation into BCI illiteracy,

    M.-H. Lee, O.-Y . Kwon, Y .-J. Kim, H.-K. Kim, Y .-E. Lee, J. Williamson, S. Fazli, and S.-W. Lee, “EEG dataset and OpenBMI toolbox for three BCI paradigms: An investigation into BCI illiteracy,”GigaScience, vol. 8, no. 5, p. giz002, 2019

    work page 2019

  39. [39]

    BCI2000: a general-purpose brain-computer interface (BCI) system,

    G. Schalk, D. J. McFarland, T. Hinterberger, N. Birbaumer, and J. R. Wolpaw, “BCI2000: a general-purpose brain-computer interface (BCI) system,”IEEE Transactions on biomedical engineering, vol. 51, no. 6, pp. 1034–1043, 2004

    work page 2004

  40. [40]

    Electroencephalogram (EEG)-based computer-aided technique to diagnose major depressive disorder (MDD),

    W. Mumtaz, L. Xia, S. S. A. Ali, M. A. M. Yasin, M. Hussain, and A. S. Malik, “Electroencephalogram (EEG)-based computer-aided technique to diagnose major depressive disorder (MDD),”Biomedical Signal Processing and Control, vol. 31, pp. 108–115, 2017

    work page 2017

  41. [41]

    A dataset of scalp EEG recordings of Alzheimer’s disease, frontotemporal dementia and healthy subjects from routine EEG,

    A. Miltiadous, K. D. Tzimourta, T. Afrantou, P. Ioannidis, N. Grigo- riadis, D. G. Tsalikakis, P. Angelidis, M. G. Tsipouras, E. Glavas, N. Giannakeaset al., “A dataset of scalp EEG recordings of Alzheimer’s disease, frontotemporal dementia and healthy subjects from routine EEG,”Data, vol. 8, no. 6, p. 95, 2023

    work page 2023

  42. [42]

    UC San Diego Resting State EEG Data from Patients with Parkinson’s Disease,

    A. P. Rockhill, N. Jackson, J. George, A. Aron, and N. C. Swann, “UC San Diego Resting State EEG Data from Patients with Parkinson’s Disease,” OpenNeuro, 2021

    work page 2021

  43. [43]

    Electroencephalograms during mental arithmetic task performance,

    I. Zyma, S. Tukaev, I. Seleznov, K. Kiyono, A. Popov, M. Chernykh, and O. Shpenkov, “Electroencephalograms during mental arithmetic task performance,”Data, vol. 4, no. 1, p. 14, 2019

    work page 2019

  44. [44]

    PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals,

    A. L. Goldberger, L. A. Amaral, L. Glass, J. M. Hausdorff, P. C. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, and H. E. Stanley, “PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals,”circulation, vol. 101, no. 23, pp. e215–e220, 2000

    work page 2000

  45. [45]

    Simultaneous acquisition of EEG and NIRS during cognitive tasks for an open access dataset,

    J. Shin, A. V on L ¨uhmann, D.-W. Kim, J. Mehnert, H.-J. Hwang, and K.- R. M¨uller, “Simultaneous acquisition of EEG and NIRS during cognitive tasks for an open access dataset,”Scientific data, vol. 5, no. 1, p. 180003, 2018

    work page 2018

  46. [46]

    MGFormer: A lightweight multi-granular transformer for subject-independent Alzheimer’s classification,

    R. Rahman, A. K. al Azad, and S. Momen, “MGFormer: A lightweight multi-granular transformer for subject-independent Alzheimer’s classification,”Biomedical Signal Processing and Control, vol. 110, p. 108230, 2025. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S1746809425007414

    work page 2025

  47. [47]

    Motor imagery and direct brain- computer communication,

    G. Pfurtscheller and C. Neuper, “Motor imagery and direct brain- computer communication,”Proceedings of the IEEE, vol. 89, no. 7, pp. 1123–1134, 2001

    work page 2001

  48. [48]

    The neurophysiological basis of motor imagery,

    J. Decety, “The neurophysiological basis of motor imagery,”Behavioural brain research, vol. 77, no. 1-2, pp. 45–52, 1996

    work page 1996

  49. [49]

    Multivariate pattern analysis of EEG-based functional connectivity: A study on the identification of depression,

    H. Peng, C. Xia, Z. Wang, J. Zhu, X. Zhang, S. Sun, J. Li, X. Huo, and X. Li, “Multivariate pattern analysis of EEG-based functional connectivity: A study on the identification of depression,”Ieee Access, vol. 7, pp. 92 630–92 641, 2019

    work page 2019

  50. [50]

    Deep-learning-based imaging-classification identified cingulate island sign in dementia with Lewy bodies,

    T. Iizuka, M. Fukasawa, and M. Kameyama, “Deep-learning-based imaging-classification identified cingulate island sign in dementia with Lewy bodies,”Scientific reports, vol. 9, no. 1, p. 8944, 2019

    work page 2019

  51. [51]

    Investigation of EEG abnormalities in the early stage of Parkinson’s disease,

    C.-X. Han, J. Wang, G.-S. Yi, and Y .-Q. Che, “Investigation of EEG abnormalities in the early stage of Parkinson’s disease,”Cognitive neurodynamics, vol. 7, no. 4, pp. 351–359, 2013

    work page 2013

  52. [52]

    EEG fingerprints of task-independent mental workload discrimination,

    I. Kakkos, G. N. Dimitrakopoulos, Y . Sun, J. Yuan, G. K. Matsopoulos, A. Bezerianos, and Y . Sun, “EEG fingerprints of task-independent mental workload discrimination,”IEEE Journal of Biomedical and Health Informatics, vol. 25, no. 10, pp. 3824–3833, 2021

    work page 2021

  53. [53]

    Top-down modulation of neural activity in anticipatory visual attention: control mechanisms revealed by simultaneous EEG-fMRI,

    Y . Liu, J. Bengson, H. Huang, G. R. Mangun, and M. Ding, “Top-down modulation of neural activity in anticipatory visual attention: control mechanisms revealed by simultaneous EEG-fMRI,”Cerebral cortex, vol. 26, no. 2, pp. 517–529, 2016

    work page 2016