pith. sign in

arxiv: 2605.20182 · v1 · pith:MH3DSQRAnew · submitted 2026-05-19 · 💻 cs.LG · cs.AI

Atoms of Thought: Universal EEG Representation Learning with Microstates

Xinyang Tian , Ruitao Liu , Ziyi Ye , Siyang Xue , Xin Wang , Xuesong Chen This is my paper

Pith reviewed 2026-05-20 06:17 UTC · model grok-4.3

classification 💻 cs.LG cs.AI
keywords EEGmicrostatesrepresentation learninguniversal tokenizerbrain-computer interfacessleep stagingemotion recognitionmotor imagery
0
0 comments X

The pith

Microstate sequences created by clustering large EEG datasets serve as universal building blocks that outperform time-domain and frequency-domain features on sleep staging, emotion recognition, and motor imagery tasks.

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

The paper tests whether continuous EEG signals can be tokenized into discrete microstates that capture core brain activity patterns. These microstates are derived once from a large medical dataset and then applied without retraining to multiple downstream problems. Results indicate that this representation improves accuracy over conventional signal features in different models. If the claim holds, brain signals would possess a reusable discrete vocabulary rather than requiring task-specific feature engineering each time.

Core claim

By clustering continuous EEG recordings from a large medical dataset into sequences of discrete microstates, the authors construct a universal tokenizer. This tokenizer converts raw signals into microstate sequences that function as input representations for downstream models. The sequences yield higher performance than time-domain or frequency-domain features across sleep staging, emotion recognition, and motor imagery classification, while also supporting greater interpretability and scalability.

What carries the argument

The microstate tokenizer, formed by clustering EEG signals into discrete states that represent fundamental short-scale brain activity patterns.

If this is right

  • Microstate sequences can be reused across unrelated EEG tasks after a single training step on the clustering dataset.
  • The discrete nature of microstates allows direct inspection of which patterns contribute to each classification decision.
  • The approach reduces the need for hand-crafted features when building models for new cognitive or clinical applications.
  • Scalability improves because the tokenizer is computed once and then applied universally rather than retrained per task.

Where Pith is reading between the lines

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

  • If microstates prove stable across populations, they could serve as reference patterns for detecting deviations linked to neurological conditions.
  • Pre-computed tokenizers might enable lighter, on-device EEG processing for real-time brain-computer interfaces.
  • The same clustering procedure could be tested on other physiological signals to check whether discrete state representations generalize beyond EEG.

Load-bearing premise

Clustering EEG signals from one large dataset produces a fixed set of microstates that remain effective across new subjects, recording conditions, and tasks without any retraining or adjustment.

What would settle it

Applying the same fixed microstate tokenizer to a new EEG dataset recorded under different hardware or subject demographics and observing that accuracy falls below task-specific time-frequency baselines.

Figures

Figures reproduced from arXiv: 2605.20182 by Ruitao Liu, Siyang Xue, Xin Wang, Xinyang Tian, Xuesong Chen, Ziyi Ye.

Figure 1
Figure 1. Figure 1: Visualization of Different Representations and Downstream Tasks. Conventional representations mainly reside in [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Pipeline of our Work. Our work can be broken into two parts. The first involves fitting a tokenizer to extract microstates [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Accuracy (left) and Cohen’s Kappa (right) with dif [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Accuracy and Cohen’s Kappa under Sleep Net Zero [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Visualizing Microstates Distribution. Visualization of the distribution of different microstates among different subjects [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The Distribution of EEG Leads Filtering and Resampling. The raw EEG signals are then bandpass filtered between 1Hz and 40Hz, followed by a resampling at 𝑓𝑟𝑒𝑠 = 100Hz. After these procedures, the raw EEG signals now have shape 𝑠𝑟𝑒𝑠 ∈ R 𝑁 ×𝑓𝑟𝑒𝑠𝑇 . Having obtained the resampled data, we construct the representations accordingly. Constructing microstates. The fitted clustering model is applied to the resampled… view at source ↗
Figure 7
Figure 7. Figure 7: The Valence-Arousal Space A.3 Motor Imagery Classification A.3.1 Dataset. We use the Motor Movement/Imagery dataset [25, 53]. The dataset consists of 109 subjects undergoing 14 trials. The 14 trials includes two rest sessions and four tasks. The four tasks are: • Task 1: Open and close the left or right fist. • Task 2: Imagine opening and closing the left or right fist. • Task 3: Open and close both fists … view at source ↗
Figure 10
Figure 10. Figure 10: Visualizing Microstates 161, 385, 419 and 421 [PITH_FULL_IMAGE:figures/full_fig_p022_10.png] view at source ↗
Figure 8
Figure 8. Figure 8: The Model Structure of ResBlocks with in=N, out=M [PITH_FULL_IMAGE:figures/full_fig_p022_8.png] view at source ↗
Figure 12
Figure 12. Figure 12: Visualizing Microstates 489 which also has relatively high potential with all leads between 2 ∼ 14𝜇V. This again shows that from W through N1, N2 to N3, high amplitude brain activity becomes more and more common. 23 [PITH_FULL_IMAGE:figures/full_fig_p023_12.png] view at source ↗
read the original abstract

Learning universal representations from electroencephalogram (EEG) signals is a cutting-edge approach in the field of neuroinformatics and brain-computer interfaces (BCIs). Conventionally, EEG is treated as a multivariate temporal signal, where time- or frequency-domain features are extracted for representation learning. This paper investigates a simple yet effective EEG representation, i.e., microstates. Microstates represent the building blocks of brain activity patterns at a microscopic time scale. We build a universal microstate tokenizer from a large medical EEG dataset by clustering continuous EEG signals into sequences of discrete microstates. The microstate tokenizer is then adopted universally across a series of downstream tasks, including sleep staging, emotion recognition, and motor imagery classification. Experimental results show that EEG representation learning with microstates outperforms traditional time-domain and frequency-domain features under different models and across different tasks. Further analysis shows that microstates offer greater interpretability and scalability, thereby opening up applications in both cognitive neuroscience and clinical research.

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 proposes representing EEG signals via microstates obtained by clustering a large medical EEG corpus into a fixed discrete tokenizer. This tokenizer produces sequences that are fed to standard models for downstream tasks including sleep staging, emotion recognition, and motor imagery classification. The central claim is that these microstate sequences outperform conventional time-domain and frequency-domain features across models and tasks while also providing greater interpretability and scalability.

Significance. If the universality claim holds with proper controls, the work would supply a discrete, parameter-light alphabet for EEG that could improve interpretability in BCI and clinical applications and reduce the need for task-specific feature engineering. The approach aligns with recent discrete-representation trends in other modalities and could enable more scalable cross-task transfer if the microstate set proves stable.

major comments (2)
  1. [Experiments] The universality claim (abstract and Experiments section) rests on a single fixed tokenizer derived from one medical dataset being applied without retraining to heterogeneous target corpora. No topographic correlation coefficients, transition-matrix KL divergences, or other stability metrics between source and target datasets are reported, leaving open the possibility that gains arise from dataset alignment rather than true cross-domain invariance.
  2. [Results] The abstract and results sections assert outperformance over time- and frequency-domain baselines but supply no quantitative numbers, error bars, dataset sizes, subject counts, or statistical tests. Without these, the support for the central claim cannot be evaluated; an ablation retraining the tokenizer per target domain is also absent.
minor comments (2)
  1. [Methods] Notation for the microstate tokenizer (e.g., definition of the clustering objective and sequence encoding) should be introduced with an equation in the Methods section for reproducibility.
  2. [Figures] Figure captions for microstate topographies and transition diagrams should explicitly state the number of clusters (K) and the source dataset used for fitting.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. We address each major comment point by point below and have revised the paper accordingly to strengthen the presentation of our results and claims.

read point-by-point responses

  1. Referee: [Experiments] The universality claim (abstract and Experiments section) rests on a single fixed tokenizer derived from one medical dataset being applied without retraining to heterogeneous target corpora. No topographic correlation coefficients, transition-matrix KL divergences, or other stability metrics between source and target datasets are reported, leaving open the possibility that gains arise from dataset alignment rather than true cross-domain invariance.

    Authors: We agree that explicit stability metrics would better support the universality claim. In the revised manuscript, we now report topographic correlation coefficients and KL divergences between transition matrices computed on the source medical EEG corpus versus each target dataset (sleep staging, emotion recognition, motor imagery). These metrics indicate strong alignment of the microstate distributions across domains, consistent with the fixed tokenizer enabling true cross-domain transfer rather than dataset-specific effects. We also clarify in the text that no retraining of the tokenizer occurs on target data. revision: yes

  2. Referee: [Results] The abstract and results sections assert outperformance over time- and frequency-domain baselines but supply no quantitative numbers, error bars, dataset sizes, subject counts, or statistical tests. Without these, the support for the central claim cannot be evaluated; an ablation retraining the tokenizer per target domain is also absent.

    Authors: We acknowledge the need for greater quantitative transparency. The revised Results section now includes specific performance metrics (accuracy and F1 scores) with error bars (standard deviation across folds or subjects), dataset sizes, subject counts, and statistical tests (paired t-tests or Wilcoxon tests with p-values) for all model-task combinations. We have also added the requested ablation: retraining the tokenizer independently on each target domain. Results show the fixed universal tokenizer performs on par or better than domain-specific versions, reinforcing the value of large-scale pretraining. revision: yes

Circularity Check

0 steps flagged

No significant circularity in microstate tokenizer construction or downstream evaluation

full rationale

The paper describes an empirical pipeline: clustering EEG signals from one large medical dataset to form a fixed discrete tokenizer, then feeding the resulting sequences into models for separate downstream tasks (sleep staging, emotion recognition, motor imagery). No equations, derivations, or first-principles claims are presented that reduce performance metrics to fitted parameters or self-referential definitions. The reported outperformance rests on direct comparisons against time- and frequency-domain baselines on external task benchmarks rather than any internal redefinition or self-citation chain. This structure is self-contained against the stated benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 1 invented entities

The approach depends on the domain assumption that microstates constitute fundamental building blocks and introduces the tokenizer as a new construct; clustering implies at least one free parameter for the number of states.

free parameters (1)
  • Number of microstates
    The number of clusters used to discretize EEG signals into the tokenizer vocabulary; this choice directly defines the representation granularity.
axioms (1)
  • domain assumption Microstates represent the building blocks of brain activity patterns at a microscopic time scale
    Invoked in the abstract as the foundational premise for treating EEG as sequences of these states rather than continuous signals.
invented entities (1)
  • Microstate tokenizer no independent evidence
    purpose: Convert continuous EEG into reusable sequences of discrete states for universal downstream application
    New tool introduced in this work; no independent evidence such as a predicted property outside the reported experiments is provided.

pith-pipeline@v0.9.0 · 5706 in / 1469 out tokens · 47765 ms · 2026-05-20T06:17:40.351091+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

92 extracted references · 92 canonical work pages · 4 internal anchors

  1. [1]

    Arshia Afzal, Grigorios Chrysos, Volkan Cevher, and Mahsa Shoaran. 2024. Rest: Efficient and accelerated eeg seizure analysis through residual state updates. arXiv preprint arXiv:2406.16906(2024)

    work page arXiv 2024

  2. [2]

    Aggarwal, Philip S

    Charu C. Aggarwal, Philip S. Yu, Jiawei Han, and Jianyong Wang. 2003. - A Frame- work for Clustering Evolving Data Streams. InProceedings 2003 VLDB Conference, Johann-Christoph Freytag, Peter Lockemann, Serge Abiteboul, Michael Carey, Patricia Selinger, and Andreas Heuer (Eds.). Morgan Kaufmann, San Francisco, 81–92. doi:10.1016/B978-012722442-8/50016-1

    work page doi:10.1016/b978-012722442-8/50016-1 2003

  3. [3]

    Aburn, and Clinton Fookes

    David Ahmedt-Aristizabal, Tharindu Fernando, Simon Denman, Lars Petersson, Matthew J. Aburn, and Clinton Fookes. 2019. Neural Memory Networks for Seizure Type Classification.2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)(2019), 569–575. https: //api.semanticscholar.org/CorpusID:210966442

    work page 2019

  4. [4]

    Aydin Akan and Ozlem Karabiber Cura. 2021. Time–frequency signal processing: Today and future.Digital Signal Processing119 (2021), 103216. doi:10.1016/j.dsp. 2021.103216

    work page doi:10.1016/j.dsp 2021

  5. [5]

    Brandon Westover, and Jimeng Sun

    Irfan Al-Hussaini, Cao Xiao, M. Brandon Westover, and Jimeng Sun. 2019. SLEEPER: interpretable Sleep staging via Prototypes from Expert Rules. arXiv:1910.06100 [cs.LG] https://arxiv.org/abs/1910.06100

    work page arXiv 2019

  6. [6]

    Ghita Amrani, Amina Adadi, Mohammed Berrada, Zouhayr Souirti, and Saïd Boujraf. 2021. EEG signal analysis using deep learning: A systematic literature review. In2021 Fifth International Conference On Intelligent Computing in Data Sciences (ICDS). 1–8. doi:10.1109/ICDS53782.2021.9626707

    work page doi:10.1109/icds53782.2021.9626707 2021

  7. [7]

    Kleanthis Avramidis. 2021. Affective Analysis and Interpretation of Brain Re- sponses to Music Stimuli

    work page 2021

  8. [8]

    Anahit Babayan, Miray Erbey, Deniz Kumral, Janis Reinelt, Andrea Reiter, Josefin Röbbig, H. Schaare, Marie Uhlig, Alfred Anwander, Pierre-Louis Bazin, Annette Horstmann, Leonie Lampe, Vadim Nikulin, Hadas Okon-Singer, Sven Preusser, André Pampel, Christiane Rohr, Julia Sacher, Angelika Thoene-Otto, and Arno Villringer. 2019. A mind-brain-body dataset of M...

    work page doi:10.1038/sdata.2018.308 2019

  9. [9]

    Dongmei Bai, Tianshuang Qiu, and Xiaobing Li. 2007. [The sample entropy and its application in EEG based epilepsy detection].Sheng wu yi xue gong cheng xue za zhi = Journal of biomedical engineering = Shengwu yixue gongchengxue zazhi24 1 (2007), 200–5. https://api.semanticscholar.org/CorpusID:21621951

    work page 2007

  10. [10]

    Juliane Britz, Dimitri Van De Ville, and Christoph M. Michel. 2010. BOLD correlates of EEG topography reveal rapid resting-state network dynamics.Neu- roImage52, 4 (2010), 1162–1170. doi:10.1016/j.neuroimage.2010.02.052

    work page doi:10.1016/j.neuroimage.2010.02.052 2010

  11. [11]

    Michel, and Helmut Laufs

    Verena Brodbeck, Alena Kuhn, Frederic von Wegner, Astrid Morzelewski, Enzo Tagliazucchi, Sergey Borisov, Christoph M. Michel, and Helmut Laufs. 2012. EEG microstates of wakefulness and NREM sleep.NeuroImage62, 3 (2012), 2129–2139. doi:10.1016/j.neuroimage.2012.05.060

    work page doi:10.1016/j.neuroimage.2012.05.060 2012

  12. [12]

    Zheng Chen, Ziwei Yang, Lingwei Zhu, Wei Chen, Toshiyo Tamura, Naoaki Ono, Md Altaf-Ul-Amin, Shigehiko Kanaya, and Ming Huang. 2023. Automated Sleep Staging via Parallel Frequency-Cut Attention.IEEE Transactions on Neural Systems and Rehabilitation Engineering31 (2023), 1974–1985. doi:10.1109/TNSRE. 2023.3243589

    work page doi:10.1109/tnsre 2023

  13. [13]

    Cheng, Hanlin Goh, Kaan Dogrusoz, Oncel Tuzel, and Erdrin Azemi

    Joseph Y. Cheng, Hanlin Goh, Kaan Dogrusoz, Oncel Tuzel, and Erdrin Azemi

    work page

  14. [14]

    https://api.semanticscholar.org/CorpusID:220425132

    Subject-Aware Contrastive Learning for Biosignals.ArXivabs/2007.04871 (2020). https://api.semanticscholar.org/CorpusID:220425132

    work page arXiv 2007

  15. [15]

    Zhuoling Cheng, Xuekui Bu, Qingnan Wang, Tao Yang, and Jihui Tu. 2024. EEG-based emotion recognition using multi-scale dynamic CNN and gated trans- former.Scientific Reports14 (2024). https://api.semanticscholar.org/CorpusID: 275117639

    work page 2024

  16. [16]

    Ming Chu and Jingfeng Bi. 2023. Six classes of motor imagery EEG signals in the upper limb. doi:10.21227/8qw6-f578

    work page doi:10.21227/8qw6-f578 2023

  17. [17]

    Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, and Yoshua Bengio. 2014. Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. arXiv:1412.3555 [cs.NE] https://arxiv.org/abs/1412.3555

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  18. [18]

    Jiaxiang Dong, Haixu Wu, Haoran Zhang, Li Zhang, Jianmin Wang, and Ming- sheng Long. 2023. SimMTM: A Simple Pre-Training Framework for Masked Time- Series Modeling. InAdvances in Neural Information Processing Systems, A. Oh, T. Naumann, A. Globerson, K. Saenko, M. Hardt, and S. Levine (Eds.), Vol. 36. Curran Associates, Inc., 29996–30025. https://proceedi...

    work page 2023

  19. [19]

    Xiaobing Du, Cuixia Ma, Guanhua Zhang, Jinyao Li, Yu-Kun Lai, Guozhen Zhao, Xiaoming Deng, Yong-Jin Liu, and Hongan Wang. 2022. An Efficient LSTM Network for Emotion Recognition From Multichannel EEG Signals.IEEE Transactions on Affective Computing13, 3 (2022), 1528–1540. doi:10.1109/TAFFC. 2020.3013711

    work page doi:10.1109/taffc 2022

  20. [20]

    Ruo-Nan Duan, Jia-Yi Zhu, and Bao-Liang Lu. 2013. Differential entropy feature for EEG-based emotion classification. In6th International IEEE/EMBS Conference on Neural Engineering (NER). IEEE, 81–84

    work page 2013

  21. [21]

    Robert Efron. 1970. The minimum duration of a perception.Neuropsychologia8, 1 (1970), 57–63. doi:10.1016/0028-3932(70)90025-4

    work page doi:10.1016/0028-3932(70)90025-4 1970

  22. [22]

    Abdeljalil El Hadiri, Lhoussain Bahatti, Abdelmounime El Magri, and Rachid Lajouad. 2024. Sleep stages detection based on analysis and optimisation of non-linear brain signal parameters.Results in Engineering23 (2024), 102664. doi:10.1016/j.rineng.2024.102664

    work page doi:10.1016/j.rineng.2024.102664 2024

  23. [23]

    MohammadReza EskandariNasab, Zahra Raeisi, Reza Ahmadi Lashaki, and Hamidreza Najafi. 2024. A GRU–CNN model for auditory attention detection using microstate and recurrence quantification analysis.Scientific Reports14 (2024). https://api.semanticscholar.org/CorpusID:269211640

    work page 2024

  24. [24]

    Shenzhi Fang, Chaofeng Zhu, Jinying Zhang, Luyan Wu, Yuying Zhang, Huapin Huang, and Wanhui Lin. 2024. EEG microstates in epilepsy with and without cognitive dysfunction: Alteration in intrinsic brain activity.Epilepsy & Behavior 154 (2024), 109729. doi:10.1016/j.yebeh.2024.109729

    work page doi:10.1016/j.yebeh.2024.109729 2024

  25. [25]

    Linda Fiorini, Francesco Bossi, and Francesco Di Gruttola. 2024. EEG-based emotional valence and emotion regulation classification: a data-centric and explainable approach.Scientific reports14, 1 (October 2024), 24046. doi:10.1038/ s41598-024-75263-x

    work page 2024

  26. [26]

    Goldberger, L

    A. Goldberger, L. Amaral, L. Glass, J. Hausdorff, P. C. Ivanov, R. Mark, J. E. Mietus, G. B. Moody, Peng C. K., and H. E. Stanley. 2000. PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals.Circulation101, 23 (2000), e215–e220. Online

    work page 2000

  27. [27]

    Xiaotong Gu, Zehong Cao, Alireza Jolfaei, Peng Xu, Dongrui Wu, Tzyy-Ping Jung, and Chin-Teng Lin. 2021. EEG-Based Brain-Computer Interfaces (BCIs): A Survey of Recent Studies on Signal Sensing Technologies and Computational Intelligence Approaches and Their Applications.IEEE/ACM Transactions on Computational Biology and Bioinformatics18, 5 (2021), 1645–16...

    work page arXiv 2021

  28. [28]

    Haokun Gui, Xiucheng Li, and Xinyang Chen. 2024. Vector Quantization Pre- training for EEG Time Series with Random Projection and Phase Alignment. In Proceedings of the 41st International Conference on Machine Learning (Proceedings of Machine Learning Research, Vol. 235), Ruslan Salakhutdinov, Zico Kolter, Kather- ine Heller, Adrian Weller, Nuria Oliver, ...

    work page 2024

  29. [29]

    Abir Hadriche, Laurent Pezard, Jean-Louis Nandrino, Hamadi Ghariani, Abden- naceur Kachouri, and Viktor K. Jirsa. 2013. Mapping the dynamic repertoire of the resting brain.NeuroImage78 (2013), 448–462. doi:10.1016/j.neuroimage.2013. 04.041

    work page doi:10.1016/j.neuroimage.2013 2013

  30. [30]

    Jingzhao Hu, Chen Wang, Qiaomei Jia, Qirong Bu, Richard Sutcliffe, and Jun Feng

    work page

  31. [31]

    Neurocomputing463 (2021), 177–184

    ScalingNet: Extracting features from raw EEG data for emotion recognition. Neurocomputing463 (2021), 177–184. doi:10.1016/j.neucom.2021.08.018

    work page doi:10.1016/j.neucom.2021.08.018 2021

  32. [32]

    Sunhee Hwang, Kibeom Hong, Guiyoung Son, and Hyeran Byun. 2020. Learning CNN features from DE features for EEG-based emotion recognition.Pattern Analysis and Applications23, 3 (2020), 1323 – 1335. doi:10.1007/s10044-019- 00860-w Cited by: 104

    work page doi:10.1007/s10044-019- 2020

  33. [33]

    Abhishek Iyer, Srimit Sritik Das, Reva Teotia, Shishir Maheshwari, and Rishi Sharma. 2022. CNN and LSTM based Ensemble Learning for Human Emotion Recognition using EEG Recordings.Multimedia Tools and Applications(04 2022). doi:10.1007/s11042-022-12310-7

    work page doi:10.1007/s11042-022-12310-7 2022

  34. [34]

    Khare, Victoria Blanes-Vidal, Esmaeil S

    Smith K. Khare, Victoria Blanes-Vidal, Esmaeil S. Nadimi, and U. Rajendra Acharya. 2024. Emotion recognition and artificial intelligence: A systematic review (2014–2023) and research recommendations.Information Fusion102 (2024), 102019. doi:10.1016/j.inffus.2023.102019

    work page doi:10.1016/j.inffus.2023.102019 2024

  35. [35]

    Domant˙e Kučikien˙e, Ravichandran Rajkumar, Katharina Timpte, Jan Heckel- mann, Irene Neuner, Yvonne Weber, and Stefan Wolking. 2024. EEG microstates show different features in focal epilepsy and psychogenic nonepileptic seizures. Epilepsia65 (01 2024). doi:10.1111/epi.17897

    work page doi:10.1111/epi.17897 2024

  36. [36]

    Byeong-Hoo Lee, Ji-Hoon Jeong, Kyung-Hwan Shim, and Dong-Joo Kim. 2020. Motor Imagery Classification of Single-Arm Tasks Using Convolutional Neural Network based on Feature Refining. arXiv:2002.01122 [cs.HC] https://arxiv.org/ abs/2002.01122

    work page arXiv 2020

  37. [37]

    Lehmann, H

    D. Lehmann, H. Ozaki, and I. Pal. 1987. EEG alpha map series: brain micro-states by space-oriented adaptive segmentation.Electroencephalography and Clinical Neurophysiology67, 3 (1987), 271–288. doi:10.1016/0013-4694(87)90025-3

    work page doi:10.1016/0013-4694(87)90025-3 1987

  38. [38]

    D Lehmann, W.K Strik, B Henggeler, T Koenig, and M Koukkou. 1998. Brain electric microstates and momentary conscious mind states as building blocks of spontaneous thinking: I. Visual imagery and abstract thoughts.International Journal of Psychophysiology29, 1 (1998), 1–11. doi:10.1016/S0167-8760(97)00098- 6

    work page doi:10.1016/s0167-8760(97)00098- 1998

  39. [39]

    Shuzhen Li, Yuxin Chen, Xuesong Chen, Ruiyang Gao, Yupeng Zhang, Chao Yu, Yunfei Li, Ziyi Ye, Weijun Huang, Hongliang Yi, et al. 2024. SleepNetZero: Zero-Burden Zero-Shot Reliable Sleep Staging with Neural Networks Based on Ballistocardiograms.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies8, 4 (2024), 1–25

    work page 2024

  40. [40]

    Robert Lin, Ren-Guey Lee, Chwan-Lu Tseng, Heng-Kuan Zhou, C. F. Chao, and Joe-Air Jiang. 2006. A NEW APPROACH FOR IDENTIFYING SLEEP AP- NEA SYNDROME USING WAVELET TRANSFORM AND NEURAL NETWORKS. Biomedical Engineering: Applications, Basis and Communications18 (2006), 138–

    work page 2006

  41. [41]

    https://api.semanticscholar.org/CorpusID:2412588

    work page

  42. [42]

    Hong Liu, Haoling Tang, Wei Wei, Gesheng Wang, Yong Du, and Jianghai Ruan

    work page

  43. [43]

    https://api.semanticscholar.org/CorpusID: 232359456

    Altered peri-seizure EEG microstate dynamics in patients with absence9 epilepsy.Seizure88 (2021), 15–21. https://api.semanticscholar.org/CorpusID: 232359456

    work page 2021

  44. [44]

    Huisheng Lu, Mingshi Wang, and Hongqiang Yu. 2005. EEG Model and Location in Brain when Enjoying Music.2005 IEEE Engineering in Medicine and Biology 27th Annual Conference(2005), 2695–2698. https://api.semanticscholar.org/CorpusID: 21783113

    work page 2005

  45. [45]

    Marzia Lucia, Christoph Michel, Stephanie Clarke, and Micah Murray. 2007. Single-subject EEG analysis based on topographic information.International Journal of Bioelectromagnetism www.ijbem.org9 (01 2007), 168–171

    work page 2007

  46. [46]

    Scott Makeig, Stefan Debener, Julie Onton, and Arnaud Delorme. 2004. Mining event-related brain dynamics.Trends in Cognitive Sciences8, 5 (2004), 204–210. doi:10.1016/j.tics.2004.03.008

    work page doi:10.1016/j.tics.2004.03.008 2004

  47. [47]

    Michel and Thomas Koenig

    Christoph M. Michel and Thomas Koenig. 2018. EEG microstates as a tool for studying the temporal dynamics of whole-brain neuronal networks: A review. NeuroImage180 (2018), 577–593. doi:10.1016/j.neuroimage.2017.11.062 Brain Connectivity Dynamics

    work page doi:10.1016/j.neuroimage.2017.11.062 2018

  48. [48]

    Milz, P.L

    P. Milz, P.L. Faber, D. Lehmann, T. Koenig, K. Kochi, and R.D. Pascual-Marqui

    work page

  49. [49]

    doi:10.1016/j.neuroimage.2015

    The functional significance of EEG microstates—Associations with modali- ties of thinking.NeuroImage125 (2016), 643–656. doi:10.1016/j.neuroimage.2015. 08.023

    work page doi:10.1016/j.neuroimage.2015 2016

  50. [50]

    Micah Murray, Denis Brunet, and Christoph Michel. 2008. Topographic ERP Analyses: A Step-by-Step Tutorial Review.Brain topography20 (07 2008), 249–64. doi:10.1007/s10548-008-0054-5

    work page doi:10.1007/s10548-008-0054-5 2008

  51. [51]

    A Time Series is Worth 64 Words: Long-term Forecasting with Transformers

    Yuqi Nie, Nam H. Nguyen, Phanwadee Sinthong, and Jayant Kalagnanam. 2022. A Time Series is Worth 64 Words: Long-term Forecasting with Transformers.ArXiv abs/2211.14730 (2022). https://api.semanticscholar.org/CorpusID:254044221

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  52. [52]

    Pascual-Marqui, C.M

    R.D. Pascual-Marqui, C.M. Michel, and D. Lehmann. 1995. Segmentation of brain electrical activity into microstates: model estimation and validation.IEEE Transactions on Biomedical Engineering42, 7 (1995), 658–665. doi:10.1109/10. 391164

    work page doi:10.1109/10 1995

  53. [53]

    Darkner, Lykke Kempfner, Miki Nikolic, Poul Jørgen Jennum, and C

    Mathias Perslev, S. Darkner, Lykke Kempfner, Miki Nikolic, Poul Jørgen Jennum, and C. Igel. 2021. U-Sleep: resilient high-frequency sleep staging.NPJ Digital Medicine4 (2021). https://api.semanticscholar.org/CorpusID:233237814

    work page 2021

  54. [54]

    Chén, Philipp Koch, Alfred Mertins, and Maarten De Vos

    Huy Phan, Kaare Mikkelsen, Oliver Y. Chén, Philipp Koch, Alfred Mertins, and Maarten De Vos. 2022. SleepTransformer: Automatic Sleep Staging With In- terpretability and Uncertainty Quantification.IEEE Transactions on Biomedical Engineering69, 8 (2022), 2456–2467. doi:10.1109/TBME.2022.3147187

    work page doi:10.1109/tbme.2022.3147187 2022

  55. [55]

    Michel, and Patrik Vuilleu- mier

    Gilles Pourtois, Sylvain Delplanque, Christoph M. Michel, and Patrik Vuilleu- mier. 2008. Beyond Conventional Event-related Brain Potential (ERP): Ex- ploring the Time-course of Visual Emotion Processing Using Topographic and Principal Component Analyses.Brain Topography20 (2008), 265–277. https://api.semanticscholar.org/CorpusID:15084282

    work page 2008

  56. [56]

    Giulia Prete, Pierpaolo Croce, Filippo Zappasodi, Luca Tommasi, and Paolo Capotosto. 2022. Exploring brain activity for positive and negative emotions by means of EEG microstates.Scientific Reports12 (03 2022), 1–11. doi:10.1038/ s41598-022-07403-0

    work page 2022

  57. [57]

    Asha S.A, Sudalaimani C, Devanand P, Alexander G, Arya Maniyan Lathikaku- mari, Sanjeev V Thomas, and Ramshekhar N Menon. 2024. Analysis of EEG mi- crostates as biomarkers in neuropsychological processes – Review.Computers in Biology and Medicine173 (2024), 108266. doi:10.1016/j.compbiomed.2024.108266

    work page doi:10.1016/j.compbiomed.2024.108266 2024

  58. [58]

    Schalk, D.J

    G. Schalk, D.J. McFarland, T. Hinterberger, N. Birbaumer, and J.R. Wolpaw. 2004. BCI2000: a general-purpose brain-computer interface (BCI) system.IEEE Trans- actions on Biomedical Engineering51, 6 (2004), 1034–1043. doi:10.1109/TBME. 2004.827072

    work page doi:10.1109/tbme 2004

  59. [59]

    Bastian Schiller, Matthias Sperl, Tobias Kleinert, Kyle Nash, and Lorena Gianotti

    work page

  60. [60]

    doi:10.1007/s10548-023-00987-4

    EEG Microstates in Social and Affective Neuroscience.Brain Topography 37 (07 2023), 1–17. doi:10.1007/s10548-023-00987-4

    work page doi:10.1007/s10548-023-00987-4 2023

  61. [61]

    Lehmann, Pascal Faber, Patricia Milz, and Lorena Gianotti

    Felix Schlegel, D. Lehmann, Pascal Faber, Patricia Milz, and Lorena Gianotti

    work page

  62. [62]

    doi:10.1007/s10548-011-0189-7

    EEG Microstates During Resting Represent Personality Differences.Brain topography25 (06 2011), 20–6. doi:10.1007/s10548-011-0189-7

    work page doi:10.1007/s10548-011-0189-7 2011

  63. [63]

    A new finite element paradigm to solve contact prob- lems with roughness

    Benjamin A. Seitzman, Malene Abell, Samuel C. Bartley, Molly A. Erickson, Amanda R. Bolbecker, and William P. Hetrick. 2017. Cognitive manipulation of brain electric microstates.NeuroImage146 (2017), 533–543. doi:10.1016/j. neuroimage.2016.10.002

    work page doi:10.1016/j 2017

  64. [64]

    Xiaorui Shao and Chang Soo Kim. 2022. A Hybrid Deep Learning Scheme for Multi-Channel Sleep Stage Classification.Computers, Materials & Continua (2022). https://api.semanticscholar.org/CorpusID:243464227

    work page 2022

  65. [65]

    Xinke Shen, Xianggen Liu, Xin Hu, Dan Zhang, and Sen Song. 2023. Contrastive Learning of Subject-Invariant EEG Representations for Cross-Subject Emotion Recognition.IEEE Transactions on Affective Computing14, 3 (2023), 2496–2511. doi:10.1109/TAFFC.2022.3164516

    work page doi:10.1109/taffc.2022.3164516 2023

  66. [66]

    In-Ho Song, Doo-Soo Lee, and Sun I Kim. 2004. Recurrence quantification analysis of sleep electoencephalogram in sleep apnea syndrome in humans.Neuroscience Letters366, 2 (2004), 148–153. doi:10.1016/j.neulet.2004.05.025

    work page doi:10.1016/j.neulet.2004.05.025 2004

  67. [67]

    Tengfei Song, Suyuan Liu, Wenming Zheng, Yuan Zong, Zhen Cui, Yang Li, and Xiaoyan Zhou. 2021. Variational Instance-Adaptive Graph for EEG Emotion Recognition.IEEE Transactions on Affective Computing14 (2021), 343–356. https: //api.semanticscholar.org/CorpusID:233621668

    work page 2021

  68. [68]

    Jianlin Su, Yu Lu, Shengfeng Pan, Ahmed Murtadha, Bo Wen, and Yunfeng Liu. 2023. RoFormer: Enhanced Transformer with Rotary Position Embedding. arXiv:2104.09864 [cs.CL] https://arxiv.org/abs/2104.09864

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  69. [69]

    Puthankattil Subha, Paul K

    D. Puthankattil Subha, Paul K. Joseph, Rajendra Acharya U., and Choo Min Lim

    work page

  70. [70]

    https://api.semanticscholar.org/CorpusID:1140473

    EEG Signal Analysis: A Survey.Journal of Medical Systems34 (2010), 195–212. https://api.semanticscholar.org/CorpusID:1140473

    work page 2010

  71. [71]

    Coulthard, Jon T

    Luke Tait, Francesco Tamagnini, George Stothart, Edoardo Barvas, Chiara Monaldini, Roberto P Frusciante, Mirco Volpini, Susanna Guttmann, Elizabeth J. Coulthard, Jon T. Brown, Nina Kazanina, and Marc Goodfellow. 2019. EEG mi- crostate complexity for aiding early diagnosis of Alzheimer’s disease.Scientific Reports10 (2019). https://api.semanticscholar.org/...

    work page 2019

  72. [72]

    and Abhijit Bhattacharyya

    Padhmashree V. and Abhijit Bhattacharyya. 2022. Human emotion recognition based on time–frequency analysis of multivariate EEG signal.Knowledge-Based Systems238 (2022), 107867. doi:10.1016/j.knosys.2021.107867

    work page doi:10.1016/j.knosys.2021.107867 2022

  73. [73]

    Vanitha and P

    V. Vanitha and P. Krishnan. 2017. Time-frequency analysis of EEG for improved classification of emotion.International Journal of Biomedical Engineering and Technology23, 2-4 (2017), 191–212. arXiv:https://www.inderscienceonline.com/doi/pdf/10.1504/IJBET.2017.082661 doi:10.1504/IJBET.2017.082661

    work page doi:10.1504/ijbet.2017.082661 2017

  74. [74]

    2019.Normal Sleep EEG

    Anna Elisabetta Vaudano, Nicoletta Azzi, and Irene Trippi. 2019.Normal Sleep EEG. Springer International Publishing, Cham, 153–175. doi:10.1007/978-3-030- 04573-9_10

    work page doi:10.1007/978-3-030- 2019

  75. [75]

    Berry, and Yogatheesan Varatharajah

    Neeraj Wagh, Jionghao Wei, Samarth Rawal, Brent M. Berry, and Yogatheesan Varatharajah. 2022. Evaluating Latent Space Robustness and Uncertainty of EEG-ML Models under Realistic Distribution Shifts. arXiv:2209.11233 [eess.SP] https://arxiv.org/abs/2209.11233

    work page arXiv 2022

  76. [76]

    Fei Wang, Shichao Wu, Weiwei Zhang, Zongfeng Xu, Yahui Zhang, Chengdong Wu, and Sonya Coleman. 2020. Emotion recognition with convolutional neural network and EEG-based EFDMs.Neuropsychologia146 (2020), 107506. doi:10. 1016/j.neuropsychologia.2020.107506

    work page arXiv 2020

  77. [77]

    Guangyu Wang, Wenchao Liu, Yuhong He, Cong Xu, Lin Ma, and Haifeng Li

    work page

  78. [78]

    InThe Thirty-eighth Annual Conference on Neural Information Processing Systems

    EEGPT: Pretrained Transformer for Universal and Reliable Representation of EEG Signals. InThe Thirty-eighth Annual Conference on Neural Information Processing Systems

    work page

  79. [79]

    Jiquan Wang, Sha Zhao, Haiteng Jiang, Shijian Li, Tao Li, and Gang Pan

    work page

  80. [80]

    arXiv:2401.05363 [eess.SP] https://arxiv.org/abs/2401.05363

    Generalizable Sleep Staging via Multi-Level Domain Alignment. arXiv:2401.05363 [eess.SP] https://arxiv.org/abs/2401.05363

    work page arXiv

Showing first 80 references.