pith. sign in

arxiv: 2605.18483 · v1 · pith:VWRSR2TKnew · submitted 2026-05-18 · 💻 cs.LG · cs.AI

Modality vs. Morphology: A Framework for Time Series Classification for Biological Signals

Jordan Tschida , Matthew Yohe , Edward Kane , Gavin Jager , Emma J. Reid , Tony G. Allen , Mark Story , Leanne Thompson
show 5 more authors
Joe Hoskins Brandon Schreiber Stan Seiferth Scott Dolvin David Cornett
This is my paper

Pith reviewed 2026-05-20 12:17 UTC · model grok-4.3

classification 💻 cs.LG cs.AI
keywords time series classificationbiological signalsmorphologymodality frameworkdeep learninginterpretabilityphysiological time series
0
0 comments X

The pith

Morphology of biological waveforms determines time series classification performance more than the choice of model.

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

This paper presents a morphology-modality framework for classifying time series from biological signals. It examines how the shape and patterns in signals from the brain, heart, muscles, and eyes guide the choice of preprocessing steps and models. The central claim is that these morphological characteristics have a stronger impact on both accuracy and the ability to understand the model's decisions than whether a traditional or deep learning model is used. This matters because it explains successes in deep learning for these tasks and suggests ways to build more effective and interpretable systems for medical and biological data analysis. The review also outlines next steps such as using morphological data augmentation.

Core claim

By analyzing electroencephalography, electromyography, electrocardiography, photoplethysmography, and ocular modalities, the framework demonstrates how morphology determines preprocessing and modeling strategies. Integrating evidence across these biological signals, the framework reveals that morphology, not model class, most strongly determines performance and interpretability.

What carries the argument

The morphology-modality framework that links waveform structures including spikes, bursts, oscillations, slow drift, and hierarchical rhythms to methodological design choices.

If this is right

  • Deep models succeed when their inductive biases align with underlying waveform dynamics.
  • Morphological data augmentation improves generalization.
  • Evaluation metrics focused on morphology enhance model assessment.
  • Preprocessing and modeling strategies should be tailored to waveform morphology across modalities.

Where Pith is reading between the lines

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

  • Explicitly incorporating morphological features into models could lead to better performance in biological signal tasks.
  • The framework might be tested on additional biological signals or even non-biological ones with similar structures to see if the principle holds.
  • Transfer learning between modalities sharing similar morphologies could be more effective than currently assumed.
  • Future benchmarks should control for morphology to isolate its effect from model class.

Load-bearing premise

The review's selection of modalities and cited studies is sufficiently representative and unbiased to support the general claim that morphology is the dominant factor across biological time series.

What would settle it

A comprehensive experiment comparing performance and interpretability of different model classes on biological signals while matching their morphologies would falsify the claim if model class shows larger effects.

Figures

Figures reproduced from arXiv: 2605.18483 by Brandon Schreiber, David Cornett, Edward Kane, Emma J. Reid, Gavin Jager, Joe Hoskins, Jordan Tschida, Leanne Thompson, Mark Story, Matthew Yohe, Scott Dolvin, Stan Seiferth, Tony G. Allen.

Figure 1
Figure 1. Figure 1: Conceptual dual-framework linking morphology-driven analysis to [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Visualization of morphologies considered in this review: [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
read the original abstract

Time series classification (TSC) of biological signals has progressed from handcrafted, modality-specific approaches to deep architectures capable of representing the diverse waveform structures of underlying physiological processes (i.e., morphology). This review introduces a unified morphology--modality framework that connects waveform structure to a methodological design, revealing how spikes, bursts, oscillations, slow drift, and hierarchical rhythms inform model design. By analyzing electroencephalography, electromyography, electrocardiography, photoplethysmography, and ocular modalities (electrooculography, pupillometry, eye-tracking), the review demonstrates how morphology determines preprocessing and modeling strategies. Integrating evidence across these biological signals, the framework reveals that morphology, not model class, most strongly determines performance and interpretability. This provides insight into why deep models succeed when their inductive biases align with underlying waveform dynamics. This review also identifies future work including morphological data augmentation and evaluation metrics to improve generalization. Together, these insights position morphology-aware modeling as a unifying principle for developing generalizable, interpretable, and physiologically meaningful TSC models across biological signals.

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 manuscript reviews time series classification (TSC) methods for biological signals (EEG, EMG, ECG, PPG, and ocular modalities). It proposes a morphology-modality framework that links waveform features such as spikes, bursts, oscillations, and slow drift to choices in preprocessing, model architecture, and interpretability. The central claim is that morphology, rather than model class, most strongly determines performance and interpretability across these signals, with implications for why deep models succeed when their inductive biases align with physiological dynamics and suggestions for future morphological data augmentation and evaluation metrics.

Significance. If the synthesis is strengthened with quantitative support, the framework could serve as a unifying principle for physiologically grounded TSC design in biology and medicine, clarifying the conditions under which particular architectures generalize and offering concrete directions for augmentation strategies and metrics that better reflect waveform structure.

major comments (2)
  1. [Abstract] Abstract and concluding synthesis: the assertion that 'morphology, not model class, most strongly determines performance and interpretability' is supported only by curated qualitative examples across modalities; no quantitative meta-analysis, variance-partitioning metrics, or explicit inclusion/exclusion criteria for the cited studies are provided to substantiate the relative explanatory power of morphology versus architecture.
  2. [Introduction / Framework description] The review's selection of modalities and studies (EEG, EMG, ECG, PPG, ocular) is presented as representative, yet without a documented search strategy or bias assessment, it is unclear whether the evidence base is exhaustive enough to support the general claim that morphology dominates across biological time series.
minor comments (2)
  1. [Framework] Notation for morphological features (e.g., 'hierarchical rhythms') could be defined more explicitly with examples or a summary table to aid readers unfamiliar with the specific signal types.
  2. [Future directions] Future-work section on morphological data augmentation would benefit from at least one concrete, implementable example or pseudocode to make the suggestion actionable.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their detailed and constructive feedback on our manuscript. We appreciate the opportunity to clarify and strengthen the presentation of our morphology-modality framework. We address the major comments below and will incorporate revisions to improve the rigor of the evidence synthesis.

read point-by-point responses

  1. Referee: [Abstract] Abstract and concluding synthesis: the assertion that 'morphology, not model class, most strongly determines performance and interpretability' is supported only by curated qualitative examples across modalities; no quantitative meta-analysis, variance-partitioning metrics, or explicit inclusion/exclusion criteria for the cited studies are provided to substantiate the relative explanatory power of morphology versus architecture.

    Authors: We agree that the central claim rests on a qualitative synthesis of representative examples rather than a formal quantitative meta-analysis. The heterogeneity of datasets, metrics, and experimental designs across the cited TSC studies makes direct variance partitioning or meta-analytic comparison difficult without introducing new assumptions. In the revised manuscript we will moderate the phrasing in the abstract and conclusion to 'our synthesis indicates that morphology is a primary determinant' and add an explicit 'Study Selection and Scope' subsection that states inclusion criteria (focus on deep TSC methods for the five modalities, emphasis on waveform features) along with a limitations paragraph acknowledging the absence of quantitative dominance testing. These changes will better bound the claim without overstating the current evidence. revision: partial

  2. Referee: [Introduction / Framework description] The review's selection of modalities and studies (EEG, EMG, ECG, PPG, ocular) is presented as representative, yet without a documented search strategy or bias assessment, it is unclear whether the evidence base is exhaustive enough to support the general claim that morphology dominates across biological time series.

    Authors: The five modalities were chosen because they collectively span the major morphological archetypes (spikes, bursts, oscillations, slow drifts, hierarchical rhythms) that appear in biological TSC literature. To address the concern we will insert a short 'Scope and Selection Rationale' paragraph that (i) explains the prevalence of these signals in recent TSC benchmarks, (ii) lists the high-level inclusion criteria used (peer-reviewed deep-learning TSC papers that report both performance and some form of interpretability analysis), and (iii) explicitly states that the review is narrative rather than systematic and therefore does not claim exhaustiveness. Potential selection biases and the resulting limits on generalizability will be noted. revision: yes

Circularity Check

0 steps flagged

Literature synthesis with no load-bearing circular derivations

full rationale

The paper is a review synthesizing evidence from EEG, EMG, ECG, PPG, and ocular signals to argue that morphology determines TSC performance and interpretability more than model class. No mathematical derivations, equations, fitted parameters, or predictions appear in the provided abstract or described structure. The central claim rests on interpretive integration of cited external studies rather than any self-referential reduction, self-citation chain, or ansatz smuggled via prior work. Per the guidelines, a literature review without fitted inputs called predictions or uniqueness theorems imported from the same authors receives a low score; the argument is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

As a review paper, the work introduces no new free parameters, mathematical axioms, or invented physical entities; it relies on standard assumptions from the machine-learning and biomedical-signal literature.

pith-pipeline@v0.9.0 · 5758 in / 1028 out tokens · 28272 ms · 2026-05-20T12:17:12.205906+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

193 extracted references · 193 canonical work pages · 1 internal anchor

  1. [1]

    A systematic review of time series classification techniques used in biomedical applications,

    W. K. Wang, I. Chen, L. Hershkovich, J. Yang, A. Shetty, G. Singh, Y . Jiang, A. Kotla, J. Z. Shang, R. Yerrabelliet al., “A systematic review of time series classification techniques used in biomedical applications,”Sensors, vol. 22, no. 20, p. 8016, 2022. [Online]. Available: https://doi.org/10.3390/s22208016

    work page doi:10.3390/s22208016 2022

  2. [2]

    Using electroencephalography to measure cognitive load,

    P. Antonenko, F. Paas, R. Grabner, and T. Van Gog, “Using electroencephalography to measure cognitive load,”Educational psychology review, vol. 22, no. 4, pp. 425–438, 2010. [Online]. Available: https://doi.org/10.1007/s10648-010-9130-y

    work page doi:10.1007/s10648-010-9130-y 2010

  3. [3]

    Biomedical signal processing and artificial intelligence in eog signals,

    A. L ´opez and F. Ferrero, “Biomedical signal processing and artificial intelligence in eog signals,” inAdvances in Non-Invasive Biomedical Signal Sensing and Processing with Machine Learning. Springer, 2023, pp. 185–206. [Online]. Available: https://doi.org/10. 1007/978-3-031-23239-8 8

    work page 2023

  4. [4]

    Bio-signals in medical applications and challenges using artificial intelligence,

    M. Swapna, U. M. Viswanadhula, R. Aluvalu, V . Vardharajan, and K. Kotecha, “Bio-signals in medical applications and challenges using artificial intelligence,”Journal of Sensor and Actuator Networks, vol. 11, no. 1, p. 17, 2022. [Online]. Available: https://doi.org/10.3390/jsan11010017

    work page doi:10.3390/jsan11010017 2022

  5. [5]

    Artifact removal in physiological signals—practices and possibilities,

    K. T. Sweeney, T. E. Ward, and S. F. McLoone, “Artifact removal in physiological signals—practices and possibilities,”IEEE transactions on information technology in biomedicine, vol. 16, no. 3, pp. 488–500,

    work page

  6. [6]

    Available: https://doi.org/10.1109/TITB.2012.2188536

    [Online]. Available: https://doi.org/10.1109/TITB.2012.2188536

    work page doi:10.1109/titb.2012.2188536 2012

  7. [7]

    Survey: Time-series data preprocessing: A survey and an empirical analysis,

    A. Tawakuli, B. Havers, V . Gulisano, D. Kaiser, and T. Engel, “Survey: Time-series data preprocessing: A survey and an empirical analysis,”Journal of Engineering Research, vol. 13, no. 2, pp. 674–711, 2025. [Online]. Available: https://doi-org.ornl.idm.oclc.org/ 10.1016/j.jer.2024.02.018 17 TABLE IV THIS TABLE PROVIDES DETAILS ON SEVERAL COMMON BENCHMARK...

    work page doi:10.1016/j.jer.2024.02.018 2025

  8. [8]

    A large comparison of normalization methods on time series,

    F. T. Lima and V . M. Souza, “A large comparison of normalization methods on time series,”Big Data Research, vol. 34, p. 100407, 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/ S2214579623000400

    work page 2023

  9. [9]

    Eeg preprocessing and denoising,

    W. Peng, “Eeg preprocessing and denoising,” inEEG Signal Processing and Feature Extraction. Springer, 2019, pp. 71–87. [Online]. Available: https://doi.org/10.1007/978-981-13-9113-2 5

    work page doi:10.1007/978-981-13-9113-2 2019

  10. [10]

    Tutorial. surface emg detection, conditioning and pre-processing: Best practices,

    R. Merletti and G. Cerone, “Tutorial. surface emg detection, conditioning and pre-processing: Best practices,”Journal of Electromyography and Kinesiology, vol. 54, p. 102440, 2020. [Online]. Available: https://doi.org/10.1016/j.jelekin.2020.102440

    work page doi:10.1016/j.jelekin.2020.102440 2020

  11. [12]

    Comparison of different signal processing methods for reducing artifacts from photoplethysmograph signal,

    K. V . P. Naraharisetti and M. Bawa, “Comparison of different signal processing methods for reducing artifacts from photoplethysmograph signal,” in2011 IEEE International conference on electro/information technology. IEEE, 2011, pp. 1–8. [Online]. Available: https: //doi.org/10.1016/j.measurement.2013.05.021

    work page doi:10.1016/j.measurement.2013.05.021 2011

  12. [13]

    Trends in eeg signal feature extraction applications,

    A. K. Singh and S. Krishnan, “Trends in eeg signal feature extraction applications,”Frontiers in Artificial Intelligence, vol. 5, p. 1072801,

    work page

  13. [14]

    Available: https://doi.org/10.3389/frai.2022.1072801

    [Online]. Available: https://doi.org/10.3389/frai.2022.1072801

    work page doi:10.3389/frai.2022.1072801 2022

  14. [15]

    Time series feature extraction on basis of scalable hypothesis tests (tsfresh – a python package),

    M. Christ, N. Braun, J. Neuffer, and A. W. Kempa-Liehr, “Time series feature extraction on basis of scalable hypothesis tests (tsfresh – a python package),”Neurocomputing, vol. 307, pp. 72–77, 2018. [Online]. Available: https://doi.org/10.1016/j.neucom.2018.03.067

    work page doi:10.1016/j.neucom.2018.03.067 2018

  15. [16]

    Tsfel: Time series feature extraction library,

    M. Barandas, D. Folgado, L. Fernandes, S. Santos, M. Abreu, P. Bota, H. Liu, T. Schultz, and H. Gamboa, “Tsfel: Time series feature extraction library,”SoftwareX, vol. 11, p. 100456, 2020

    work page 2020

  16. [17]

    Feature-based information retrieval of multimodal biosignals with a self-similarity matrix: Focus on automatic segmentation,

    J. Rodrigues, H. Liu, D. Folgado, D. Belo, T. Schultz, and H. Gamboa, “Feature-based information retrieval of multimodal biosignals with a self-similarity matrix: Focus on automatic segmentation,”Biosensors, vol. 12, no. 12, p. 1182, 2022. [Online]. Available: https://doi.org/10.3390/bios12121182

    work page doi:10.3390/bios12121182 2022

  17. [18]

    Pca and ica processing methods for removal of artifacts and noise in electrocardiograms: A survey and comparison,

    M. Chawla, “Pca and ica processing methods for removal of artifacts and noise in electrocardiograms: A survey and comparison,”Applied Soft Computing, vol. 11, no. 2, pp. 2216–2226, 2011. [Online]. Available: https://doi.org/10.1016/j.asoc.2010.08.001

    work page doi:10.1016/j.asoc.2010.08.001 2011

  18. [19]

    Extracting and composing robust features with denoising autoencoders,

    P. Vincent, H. Larochelle, Y . Bengio, and P.-A. Manzagol, “Extracting and composing robust features with denoising autoencoders,” in Proceedings of the 25th international conference on Machine learning, 2008, pp. 1096–1103. [Online]. Available: https://doi.org/10. 1145/1390156.1390294

    work page arXiv 2008

  19. [20]

    Adaptive dictionary reconstruction for compressed sensing of ecg signals,

    D. Craven, B. McGinley, L. Kilmartin, M. Glavin, and E. Jones, “Adaptive dictionary reconstruction for compressed sensing of ecg signals,”IEEE journal of biomedical and health informatics, vol. 21, no. 3, pp. 645–654, 2016. [Online]. Available: https: //doi.org/10.1109/JBHI.2016.2531182

    work page doi:10.1109/jbhi.2016.2531182 2016

  20. [21]

    Developing multi-component dictionary-based sparse representation for automatic detection of epileptic eeg spikes,

    Y . Jiang, W. Chen, T. Zhang, M. Li, Y . You, and X. Zheng, “Developing multi-component dictionary-based sparse representation for automatic detection of epileptic eeg spikes,”Biomedical Signal Processing and Control, vol. 60, p. 101966, 2020. [Online]. Available: https://doi.org/10.1016/j.bspc.2020.101966

    work page doi:10.1016/j.bspc.2020.101966 2020

  21. [22]

    Rocket: exceptionally fast and accurate time series classification using random convolutional kernels,

    A. Dempster, F. Petitjean, and G. I. Webb, “Rocket: exceptionally fast and accurate time series classification using random convolutional kernels,”Data Mining and Knowledge Discovery, vol. 34, no. 5, 18 p. 1454–1495, 9 2020. [Online]. Available: http://dx.doi.org/10.1007/ s10618-020-00701-z

    work page 2020

  22. [23]

    Minirocket: A very fast (almost) deterministic transform for time series classification,

    A. Dempster, D. F. Schmidt, and G. I. Webb, “Minirocket: A very fast (almost) deterministic transform for time series classification,” in Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, ser. KDD ’21. New York, NY , USA: Association for Computing Machinery, 2021, p. 248–257. [Online]. Available: https://doi.org/10.1145/3...

    work page doi:10.1145/3447548.3467231 2021

  23. [24]

    Multirocket: multiple pooling operators and transformations for fast and effective time series classification,

    C. W. Tan, A. Dempster, C. Bergmeir, and G. I. Webb, “Multirocket: multiple pooling operators and transformations for fast and effective time series classification,”Data Min. Knowl. Discov., vol. 36, no. 5, p. 1623–1646, Sep. 2022. [Online]. Available: https://doi.org/10.1007/s10618-022-00844-1

    work page doi:10.1007/s10618-022-00844-1 2022

  24. [25]

    Classification of raw meg/eeg data with detach-rocket ensemble: an improved rocket algorithm for multivariate time series analysis,

    A. Solana, E. Frans ´en, and G. Uribarri, “Classification of raw meg/eeg data with detach-rocket ensemble: an improved rocket algorithm for multivariate time series analysis,” inInternational Workshop on Advanced Analytics and Learning on Temporal Data. Springer, 2024, pp. 96–114. [Online]. Available: https: //doi.org/10.1007/978-3-031-77066-1 6

    work page doi:10.1007/978-3-031-77066-1 2024

  25. [26]

    Comparative analysis of rocket-driven and classic eeg features in predicting attachment styles,

    D. Mizrahi, I. Laufer, and I. Zuckerman, “Comparative analysis of rocket-driven and classic eeg features in predicting attachment styles,” BMC psychology, vol. 12, no. 1, p. 87, 2024. [Online]. Available: https://doi.org/10.1186/s40359-024-01576-1

    work page doi:10.1186/s40359-024-01576-1 2024

  26. [27]

    Filter methods for feature selection – a comparative study,

    N. S ´anchez-Maro˜no, A. Alonso-Betanzos, and M. Tombilla-Sanrom ´an, “Filter methods for feature selection – a comparative study,” in Intelligent Data Engineering and Automated Learning - IDEAL 2007, H. Yin, P. Tino, E. Corchado, W. Byrne, and X. Yao, Eds. Berlin, Heidelberg: Springer Berlin Heidelberg, 2007, p. 178–187. [Online]. Available: https://doi....

    work page doi:10.1007/978-3-540-77226-2 2007

  27. [28]

    Mann–whitney u test,

    P. E. McKnight and J. Najab, “Mann–whitney u test,”The SAGE Encyclopedia of Research Design, 2022. [Online]. Available: https://api.semanticscholar.org/CorpusID:118856424

    work page 2022

  28. [29]

    Complex pearson correlation coefficient for eeg connectivity analysis,

    Z. ˇSverko, M. Vranki ´c, S. Vlahini ´c, and P. Rogelj, “Complex pearson correlation coefficient for eeg connectivity analysis,”Sensors, vol. 22, no. 4, p. 1477, 2022. [Online]. Available: https://doi.org/10.3390/ s22041477

    work page 2022

  29. [30]

    Analysis of variance: The fundamental concepts,

    S. Sawyer, “Analysis of variance: The fundamental concepts,”Journal of Manual & Manipulative Therapy, vol. 17, pp. 27E–38E, 04 2009. [Online]. Available: https::/doi.org/10.1179/jmt.2009.17.2.27E

    work page doi:10.1179/jmt.2009.17.2.27e 2009

  30. [31]

    A practical approach to feature selection,

    K. Kira and L. A. Rendell, “A practical approach to feature selection,” inMachine Learning Proceedings 1992, D. Sleeman and P. Edwards, Eds. San Francisco (CA): Morgan Kaufmann, 1992, pp. 249–256. [Online]. Available: https://www.sciencedirect.com/science/article/pii/ B9781558602472500371

    work page 1992

  31. [32]

    Overcoming the myopia of inductive learning algorithms with relieff,

    I. Kononenko, E. ˇSimec, and M. Robnik- ˇSikonja, “Overcoming the myopia of inductive learning algorithms with relieff,”Applied Intelligence, vol. 7, no. 1, pp. 39–55, 1997. [Online]. Available: https://doi.org/10.1023/A:1008280620621

    work page doi:10.1023/a:1008280620621 1997

  32. [33]

    Dna visual and analytic data mining,

    P. Hoffman, G. Grinstein, K. Marx, I. Grosse, and E. Stanley, “Dna visual and analytic data mining,” inProceedings. Visualization ’97 (Cat. No. 97CB36155), 1997, pp. 437–441. [Online]. Available: https://doi.org/10.1109/VISUAL.1997.663916

    work page doi:10.1109/visual.1997.663916 1997

  33. [34]

    Genetic algorithm for the optimization of features and neural networks in ecg signals classification,

    H. Li, D. Yuan, X. Ma, D. Cui, and L. Cao, “Genetic algorithm for the optimization of features and neural networks in ecg signals classification,”Scientific Reports, vol. 7, no. 1, p. 41011, jan 2017. [Online]. Available: https://doi.org/10.1038/srep41011

    work page doi:10.1038/srep41011 2017

  34. [35]

    Review on wrapper feature selection approaches,

    N. El Aboudi and L. Benhlima, “Review on wrapper feature selection approaches,” in2016 International Conference on Engineering & MIS (ICEMIS), 2016, pp. 1–5. [Online]. Available: https: //doi.org/10.1109/ICEMIS.2016.7745366

    work page doi:10.1109/icemis.2016.7745366 2016

  35. [36]

    Feature selection in healthcare datasets: Towards a generalizable solution,

    I. Maruotto, F. K. Ciliberti, P. Gargiulo, and M. Recenti, “Feature selection in healthcare datasets: Towards a generalizable solution,” Computers in Biology and Medicine, vol. 196, p. 110812, 9

    work page

  36. [37]

    Available: https://doi.org/10.1016/j.compbiomed.2025

    [Online]. Available: https://doi.org/10.1016/j.compbiomed.2025. 110812

    work page doi:10.1016/j.compbiomed.2025 2025

  37. [38]

    A feature selection method for multi-dimension time-series data,

    B. Kathirgamanathan and P. Cunningham, “A feature selection method for multi-dimension time-series data,” inAdvanced Analytics and Learning on Temporal Data, V . Lemaire, S. Malinowski, A. Bagnall, T. Guyet, R. Tavenard, and G. Ifrim, Eds. Cham: Springer International Publishing, 2020, p. 220–231. [Online]. Available: https://doi.org/10.1007/978-3-030-65742-0 15

    work page doi:10.1007/978-3-030-65742-0 2020

  38. [39]

    Data augmentation for time-series classification: An extensive empirical study and comprehensive survey,

    Z. Gao, H. Liu, and L. Li, “Data augmentation for time-series classification: An extensive empirical study and comprehensive survey,” Journal of Artificial Intelligence Research, vol. 83, 2025, article No. 17084. [Online]. Available: https://doi.org/10.1613/jair.1.17084

    work page doi:10.1613/jair.1.17084 2025

  39. [40]

    Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks,

    T. T. Um, F. M. J. Pfister, D. Pichler, S. Endo, M. Lang, S. Hirche, U. Fietzek, and D. Kuli ´c, “Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks,” inProceedings of the 19th ACM International Conference on Multimodal Interaction, ser. ICMI ’17. New York, NY , USA: Association for Computing M...

    work page doi:10.1145/3136755.3136817 2017

  40. [41]

    Data Augmentation for Time Series Classification using Convolutional Neural Networks,

    A. Le Guennec, S. Malinowski, and R. Tavenard, “Data Augmentation for Time Series Classification using Convolutional Neural Networks,” inECML/PKDD Workshop on Advanced Analytics and Learning on Temporal Data, Riva Del Garda, Italy, Sep. 2016. [Online]. Available: https://shs.hal.science/halshs-01357973

    work page 2016

  41. [42]

    Self-supervised contrastive pre-training for time series via time-frequency consistency,

    X. Zhang, Z. Zhao, T. Tsiligkaridis, and M. Zitnik, “Self-supervised contrastive pre-training for time series via time-frequency consistency,” inAdvances in Neural Information Processing Systems, S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh, Eds., vol. 35. Curran Associates, Inc., 2022, pp. 3988–4003. [Online]. Available: https://proc...

    work page 2022

  42. [43]

    Time series data augmentation for neural networks by time warping with a discriminative teacher,

    B. Iwana and S. Uchida, “Time series data augmentation for neural networks by time warping with a discriminative teacher,” in2020 25th International Conference on Pattern Recognition (ICPR). IEEE Computer Society, 01 2021, pp. 3558–3565. [Online]. Available: https://doi.org/10.1109/ICPR48806.2021.9412812

    work page doi:10.1109/icpr48806.2021.9412812 2021

  43. [44]

    A pilot study on the use of generative adversarial networks for data augmentation of time series,

    N. Morizet, M. Rizzato, D. Grimbert, and G. Luta, “A pilot study on the use of generative adversarial networks for data augmentation of time series,”AI, vol. 3, no. 4, pp. 789–795, 2022. [Online]. Available: https://doi.org/10.3390/ai3040047

    work page doi:10.3390/ai3040047 2022

  44. [45]

    Tsevo: Evolutionary counterfactual explanations for time series classification,

    J. H ¨ollig, C. Kulbach, and S. Thoma, “Tsevo: Evolutionary counterfactual explanations for time series classification,” in2022 21st IEEE International Conference on Machine Learning and Applications (ICMLA), 2022, pp. 29–36. [Online]. Available: https: //doi.org/10.1109/ICMLA55696.2022.00013

    work page doi:10.1109/icmla55696.2022.00013 2022

  45. [46]

    Classification of biological signals using linear and nonlinear features,

    T. Balli and R. Palaniappan, “Classification of biological signals using linear and nonlinear features,”Physiological measurement, vol. 31, no. 7, p. 903, 2010. [Online]. Available: https://doi.org/10.1088/ 0967-3334/31/7/003

    work page 2010

  46. [47]

    knn and svm classification for eeg: a review,

    M. Sha’Abani, N. Fuad, N. Jamal, and M. Ismail, “knn and svm classification for eeg: a review,” inInECCE2019: proceedings of the 5th international conference on electrical, control & computer engineering, kuantan, pahang, Malaysia, 29th July 2019. Springer, 2020, pp. 555–565. [Online]. Available: https://doi.org/10.1007/ 978-981-15-2317-5 47

    work page 2019

  47. [48]

    Comparing a supervised and an unsupervised classification method for burst detection in neonatal eeg,

    J. Lofhede, J. Degerman, N. Lofgren, M. Thordstein, A. Flisberg, I. Kjellmer, and K. Lindecrantz, “Comparing a supervised and an unsupervised classification method for burst detection in neonatal eeg,” in2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2008, pp. 3836–3839. [Online]. Available: https://doi....

    work page doi:10.1109/iembs.2008.4650046 2008

  48. [49]

    An automated, machine learning–based detection algorithm for spike- wave discharges (swds) in a mouse model of absence epilepsy,

    J. A. Pfammatter, R. K. Maganti, and M. V . Jones, “An automated, machine learning–based detection algorithm for spike- wave discharges (swds) in a mouse model of absence epilepsy,” Epilepsia Open, vol. 4, no. 1, pp. 110–122, 2019. [Online]. Available: https://doi.org/10.1002/epi4.12303

    work page doi:10.1002/epi4.12303 2019

  49. [50]

    Segmentation of the ecg signal by means of a linear regression algorithm,

    J. Aspuru, A. Ochoa-Brust, R. A. F ´elix, W. Mata-L ´opez, L. J. Mena, R. Ostos, and R. Mart ´ınez-Pel´aez, “Segmentation of the ecg signal by means of a linear regression algorithm,”Sensors, vol. 19, no. 4, p. 775, 2019. [Online]. Available: https://doi.org/10.3390/s19040775

    work page doi:10.3390/s19040775 2019

  50. [51]

    Continuous ppg-based blood pressure monitoring using multi-linear regression,

    S. Haddad, A. Boukhayma, and A. Caizzone, “Continuous ppg-based blood pressure monitoring using multi-linear regression,”IEEE journal of biomedical and health informatics, vol. 26, no. 5, pp. 2096–2105,

    work page 2096

  51. [52]

    Available: https://doi.org/10.1109/JBHI.2021.3128229

    [Online]. Available: https://doi.org/10.1109/JBHI.2021.3128229

    work page doi:10.1109/jbhi.2021.3128229 2021

  52. [53]

    Ecg signal diagnosis using discrete wavelet transform and k-nearest neighbor classifier

    Y . Toulni, T. Belhoussine Drissi, and B. Nsiri, “Ecg signal diagnosis using discrete wavelet transform and k-nearest neighbor classifier.” inProceedings of the 4th International Conference on Networking, Information Systems & Security, 2021, pp. 1–6. [Online]. Available: https://doi.org/10.1145/3454127.3457628

    work page doi:10.1145/3454127.3457628 2021

  53. [54]

    Qrs detection using k- nearest neighbor algorithm (knn) and evaluation on standard ecg databases,

    I. Saini, D. Singh, and A. Khosla, “Qrs detection using k- nearest neighbor algorithm (knn) and evaluation on standard ecg databases,”Journal of advanced research, vol. 4, no. 4, pp. 331–344,

    work page

  54. [55]

    Self-supervised Natural Image Reconstruction and Large-scale Semantic Classification from Brain Activity.NeuroImage, 254:119121, July 2022

    [Online]. Available: https://doi-org.ornl.idm.oclc.org/10.1016/j. jare.2012.05.007

    work page doi:10.1016/j 2012

  55. [56]

    Random forest based blood pressure prediction model from ecg and ppg signal,

    H. Wang, “Random forest based blood pressure prediction model from ecg and ppg signal,” inProceedings of the 2022 12th International Conference on Bioscience, Biochemistry and Bioinformatics, ser. ICBBB ’22. New York, NY , USA: Association for Computing Machinery, 2022, p. 1–6. [Online]. Available: https://doi.org/10.1145/ 3510427.3510428 19

    work page arXiv 2022

  56. [57]

    Towards quantitative precision for ecg analysis: Leveraging state space models, self-supervision and patient metadata,

    T. Mehari and N. Strodthoff, “Towards quantitative precision for ecg analysis: Leveraging state space models, self-supervision and patient metadata,”IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 11, pp. 5326–5334, 2023. [Online]. Available: https://doi.org/10.1109/JBHI.2023.3310989

    work page doi:10.1109/jbhi.2023.3310989 2023

  57. [58]

    Rhythmmamba: Fast, lightweight, and accurate remote physiological measurement,

    B. Zou, Z. Guo, X. Hu, and H. Ma, “Rhythmmamba: Fast, lightweight, and accurate remote physiological measurement,” in Proceedings of the AAAI Conference on Artificial Intelligence, no. 10. AAAI, April 2025, pp. 11 077–11 085. [Online]. Available: https://doi.org/10.1609/aaai.v39i10.33204

    work page doi:10.1609/aaai.v39i10.33204 2025

  58. [59]

    Granger causality inference in eeg source connectivity analysis: a state-space approach,

    P. Manomaisaowapak, A. Nartkulpat, and J. Songsiri, “Granger causality inference in eeg source connectivity analysis: a state-space approach,”IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 7, pp. 3146–3156, 2021. [Online]. Available: https://doi.org/10.1109/TNNLS.2021.3096642

    work page doi:10.1109/tnnls.2021.3096642 2021

  59. [60]

    Simultaneous and continuous estimation of joint angles based on surface electromyography state-space model,

    X. Xi, W. Jiang, X. Hua, H. Wang, C. Yang, Y .-B. Zhao, S. M. Miran, and Z. Luo, “Simultaneous and continuous estimation of joint angles based on surface electromyography state-space model,”IEEE Sensors Journal, vol. 21, no. 6, pp. 8089–8099, 2021. [Online]. Available: https://doi.org/10.1109/JSEN.2020.3048983

    work page doi:10.1109/jsen.2020.3048983 2021

  60. [61]

    Detection of eeg burst-suppression in neurocritical care patients using an unsupervised machine learning algorithm,

    G. Narula, M. Haeberlin, J. Balsiger, C. Str ¨assle, L. Imbach, and E. Keller, “Detection of eeg burst-suppression in neurocritical care patients using an unsupervised machine learning algorithm,”Clinical Neurophysiology, vol. 132, no. 10, pp. 2485–2492, 2021. [Online]. Available: https://doi.org/10.1016/j.clinph.2021.07.018

    work page doi:10.1016/j.clinph.2021.07.018 2021

  61. [62]

    Ecg signal reconstruction from ppg using a hybrid attention-based deep learning network,

    A. Ezzat, O. A. Omer, U. S. Mohamed, and A. S. Mubarak, “Ecg signal reconstruction from ppg using a hybrid attention-based deep learning network,”EURASIP Journal on Advances in Signal Processing, vol. 2024, no. 1, p. 95, 2024. [Online]. Available: https://doi.org/10.1186/s13634-024-01158-8

    work page doi:10.1186/s13634-024-01158-8 2024

  62. [63]

    Emmixformer: Mix transformer for eye movement recognition,

    H. Qin, H. Zhu, X. Jin, Q. Song, M. A. El-Yacoubi, and X. Gao, “Emmixformer: Mix transformer for eye movement recognition,”IEEE Transactions on Instrumentation and Measurement, 2025. [Online]. Available: https://doi.org/10.1109/TIM.2025.3551452

    work page doi:10.1109/tim.2025.3551452 2025

  63. [64]

    Hierarchical domain adaptation projective dictionary pair learning model for eeg classification in iomt systems,

    W. Cai, M. Gao, Y . Jiang, X. Gu, X. Ning, P. Qian, and T. Ni, “Hierarchical domain adaptation projective dictionary pair learning model for eeg classification in iomt systems,”IEEE Transactions on Computational Social Systems, vol. 10, no. 4, pp. 1559–1567, 2023. [Online]. Available: https://doi.org/10.1109/TCSS.2022.3176656

    work page doi:10.1109/tcss.2022.3176656 2023

  64. [65]

    Classification of electrocardiogram signals with waveform morphological analysis and support vector machines,

    H. Li, Z. An, S. Zuo, W. Zhu, L. Cao, Y . Mu, W. Song, Q. Mao, Z. Zhang, E. Liet al., “Classification of electrocardiogram signals with waveform morphological analysis and support vector machines,”Medical & Biological Engineering & Computing, vol. 60, no. 1, pp. 109–119, 2022. [Online]. Available: https: //doi.org/10.1007/s11517-021-02461-4

    work page doi:10.1007/s11517-021-02461-4 2022

  65. [66]

    Convolutional neural network based on photoplethysmography signals for sleep apnea syndrome detection,

    X. Jiang, Y . Ren, H. Wu, Y . Li, and F. Liu, “Convolutional neural network based on photoplethysmography signals for sleep apnea syndrome detection,”Frontiers in Neuroscience, vol. V olume 17 - 2023,

    work page 2023

  66. [67]

    Available: https://doi.org/10.3389/fnins.2023.1222715

    [Online]. Available: https://doi.org/10.3389/fnins.2023.1222715

    work page doi:10.3389/fnins.2023.1222715 2023

  67. [68]

    Eeg-ssm: Leveraging state-space model for dementia detection

    X.-T. Tran, L. Le, Q. T. Nguyen, T. Do, and C.-T. Lin, “Eeg-ssm: Leveraging state-space model for dementia detection,” 2024. [Online]. Available: https://arxiv.org/abs/2407.17801

    work page arXiv 2024

  68. [69]

    A robust machine learning based framework for the automated detection of adhd using pupillometric biomarkers and time series analysis,

    W. Das and S. Khanna, “A robust machine learning based framework for the automated detection of adhd using pupillometric biomarkers and time series analysis,”Scientific Reports, vol. 11, p. 16370, 8 2021. [Online]. Available: https://doi.org/10.1038/s41598-021-95673-5

    work page doi:10.1038/s41598-021-95673-5 2021

  69. [70]

    Assessing perceptual load and cognitive load by fixation-related information of eye movements,

    J.-C. Liu, K.-A. Li, S.-L. Yeh, and C. Shao-Yi, “Assessing perceptual load and cognitive load by fixation-related information of eye movements,”Sensors, vol. 22, no. 3, p. 1187, 2022. [Online]. Available: https://doi.org/10.3390/s22031187

    work page doi:10.3390/s22031187 2022

  70. [71]

    Optimal processing of surface facial emg to identify emotional expressions: A data-driven approach,

    J. M. Rutkowska, T. Ghilardi, S. V . Vacaru, J. E. van Schaik, M. Meyer, S. Hunnius, and R. Oostenveld, “Optimal processing of surface facial emg to identify emotional expressions: A data-driven approach,”Be- havior Research Methods, vol. 56, no. 7, pp. 7331–7344, 2024

    work page 2024

  71. [72]

    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. [Online]. Available: https://doi.org/10.1109/TNSRE. 2022.3230250

    work page doi:10.1109/tnsre 2023

  72. [73]

    Eegformer: A transformer–based brain activity classification method using eeg signal,

    Z. Wan, M. Li, S. Liu, J. Huang, H. Tan, and W. Duan, “Eegformer: A transformer–based brain activity classification method using eeg signal,”Frontiers in Neuroscience, vol. V olume 17 - 2023, 2023. [Online]. Available: https://doi.org/10.3389/fnins.2023.1148855

    work page doi:10.3389/fnins.2023.1148855 2023

  73. [74]

    Leveraging classical methods for accurate epilepsy detection with eeg signals,

    M. C. Karatas, S. I. Sargin, and A. J. Farmand, “Leveraging classical methods for accurate epilepsy detection with eeg signals,” in2025 33rd Signal Processing and Communications Applications Conference (SIU), 2025, pp. 1–4. [Online]. Available: https: //doi.org/10.1109/SIU66497.2025.11112105

    work page doi:10.1109/siu66497.2025.11112105 2025

  74. [75]

    Cnn-lstm and post-processing for emg-based hand gesture recognition,

    L. I. B. L ´opez, F. M. Ferri, J. Zea, ´A. L. V . Caraguay, and M. E. Benalc´azar, “Cnn-lstm and post-processing for emg-based hand gesture recognition,”Intelligent Systems with Applications, vol. 22, p. 200352,

    work page

  75. [76]

    Available: https://doi.org/10.1016/j.iswa.2024.200352

    [Online]. Available: https://doi.org/10.1016/j.iswa.2024.200352

    work page doi:10.1016/j.iswa.2024.200352 2024

  76. [77]

    E2cnn: An efficient concatenated cnn for classification of surface emg extracted from upper limb,

    M. F. Qureshi, Z. Mushtaq, M. Z. U. Rehman, and E. N. Kamavuako, “E2cnn: An efficient concatenated cnn for classification of surface emg extracted from upper limb,”IEEE Sensors Journal, vol. 23, no. 8, pp. 8989–8996, 2023. [Online]. Available: https: //doi.org/10.1109/JSEN.2023.3255408

    work page doi:10.1109/jsen.2023.3255408 2023

  77. [78]

    Transformer-based hand gesture recognition from instantaneous to fused neural decomposition of high-density emg signals,

    M. Montazerin, E. Rahimian, F. Naderkhani, S. F. Atashzar, S. Yanushkevich, and A. Mohammadi, “Transformer-based hand gesture recognition from instantaneous to fused neural decomposition of high-density emg signals,”Scientific reports, vol. 13, no. 1, p. 11000, 2023. [Online]. Available: https://doi.org/10.1038/s41598-023-36490-w

    work page doi:10.1038/s41598-023-36490-w 2023

  78. [79]

    Emg-transnn-mha: A transformer-based model for enhanced motor intent recognition in assistive robotics,

    J. A. Joby, P. Sikorski, T. Sultan, H. Akbarpour, F. Esposito, and M. Babaiasl, “Emg-transnn-mha: A transformer-based model for enhanced motor intent recognition in assistive robotics,” in2024 IEEE International Conference on Big Data (BigData). IEEE, 2024, pp. 8691–8693. [Online]. Available: https://doi.org/10.1109/ BigData62323.2024.10825417

    work page arXiv 2024

  79. [80]

    Human hand movement recognition using infinite hidden markov model based semg classification,

    R. Wen, Q. Wang, and Z. Li, “Human hand movement recognition using infinite hidden markov model based semg classification,” Biomedical Signal Processing and Control, vol. 68, p. 102592, 2021. [Online]. Available: https://doi.org/10.1016/j.bspc.2021.102592

    work page doi:10.1016/j.bspc.2021.102592 2021

  80. [81]

    Dual transformer network for predicting joint angles and torques from multi-channel emg signals in the lower limbs,

    Z. Wang, C. Chen, H. Chen, Y . Zhou, X. Wang, and X. Wu, “Dual transformer network for predicting joint angles and torques from multi-channel emg signals in the lower limbs,”IEEE Journal of Biomedical and Health Informatics, 2025. [Online]. Available: https://doi.org/10.1109/JBHI.2025.3555255

    work page doi:10.1109/jbhi.2025.3555255 2025

Showing first 80 references.