pith. sign in

arxiv: 2606.22258 · v1 · pith:6ZPUCZCJnew · submitted 2026-06-20 · 💻 cs.LG · cs.AI· eess.SP

From Handcrafted Features to Functional Edge Learning: Evolution of EEG Seizure Detection Frameworks

Sepideh Kheirollahi , Mohammad Rasoul Roshanshah This is my paper

Pith reviewed 2026-06-26 11:46 UTC · model grok-4.3

classification 💻 cs.LG cs.AIeess.SP
keywords EEG seizure detectionKolmogorov-Arnold NetworksDeep learning limitationsInterpretability in AIEpilepsy diagnosisNeural network architecturesMedical signal processing
0
0 comments X

The pith

Kolmogorov-Arnold Networks enable interpretable and efficient EEG seizure detection by using learnable functions on network connections instead of fixed activations.

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

This review examines the limitations of standard deep learning models for EEG-based seizure detection, which include opacity, need for large data, and high computation. It proposes Kolmogorov-Arnold Networks as a solution that replaces fixed neuron activations with learnable functions along edges. The paper argues this provides better parameter efficiency, inherent interpretability, and performance with scarce data. A sympathetic reader would care because it could allow transparent, deployable systems for clinical use in epilepsy monitoring.

Core claim

By replacing the fixed activation functions of traditional neurons with flexible, learnable functions along the network's connections, KANs bridge the critical gap between predictive accuracy and mathematical transparency for EEG seizure detection.

What carries the argument

Kolmogorov-Arnold Networks (KANs), which place learnable functions on the connections between neurons rather than fixed activations at nodes, enabling mathematical transparency and efficiency.

If this is right

  • Standard DL models' black-box nature is resolved, allowing physician trust in clinical settings.
  • KANs require fewer parameters, making them suitable for resource-constrained devices like wearables.
  • Performance remains robust even with limited annotated EEG data.
  • Next-generation patient-specific transparent EEG monitoring systems become feasible.

Where Pith is reading between the lines

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

  • Clinicians might prefer KAN-based models for regulatory approval due to interpretability.
  • Integration with existing EEG hardware could accelerate adoption in hospitals.
  • Future work could test KANs on real-time seizure prediction tasks beyond detection.

Load-bearing premise

That the benefits of KANs seen in general domains will apply directly to EEG seizure detection without needing major adaptations or further testing.

What would settle it

A direct comparison experiment showing that a KAN model does not outperform or match a standard DL model in accuracy or interpretability on EEG seizure datasets would falsify the central claim.

Figures

Figures reproduced from arXiv: 2606.22258 by Mohammad Rasoul Roshanshah, Sepideh Kheirollahi.

Figure 1
Figure 1. Figure 1: (a) Illustration of scalp EEG acquisition and its main frequency bands: delta, theta, alpha, beta, and gamma. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overall pipeline for data classification, progressing from dataset preparation and preprocessing to feature [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Relative proportions of published research methodologies for EEG-based seizure detection (2016–2026). [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: A conceptual network map illustrating the interconnected relationships between KANs and the prevailing [PITH_FULL_IMAGE:figures/full_fig_p021_4.png] view at source ↗
read the original abstract

Electroencephalogram (EEG) analysis remains the clinical gold standard for epilepsy diagnosis and seizure detection. While Deep Learning (DL) has significantly advanced automated EEG interpretation, its transition from controlled experimental settings to routine clinical deployment is severely bottlenecked by fundamental architectural flaws. Standard DL models operate as opaque black-boxes lacking clinical interpretability, demand massive amounts of balanced annotated data, and incur steep computational costs incompatible with resource-constrained wearable or implantable neuromodulation devices. This paper presents a comprehensive review of these prevailing limitations and explores Kolmogorov-Arnold Networks (KANs) as a emerging paradigm for EEG-based seizure detection. By replacing the fixed activation functions of traditional neurons with flexible, learnable functions along the network's connections, KANs bridge the critical gap between predictive accuracy and mathematical transparency. We systematically analyze how KAN architectures resolve the shortcomings of traditional DL-based models by offering exceptional parameter efficiency, inherent interpretability for physician trust, and robust performance under data scarcity. Ultimately, this review establishes KANs not merely as an incremental algorithmic update, but as a fundamental paradigm shift necessary to actualize next-generation, patient-specific, and thoroughly transparent clinical EEG monitoring systems.

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

1 major / 0 minor

Summary. The manuscript is a review tracing the evolution of EEG seizure detection from handcrafted features through deep learning (DL) models. It identifies DL limitations including black-box opacity, requirements for large balanced datasets, and high computational costs unsuitable for wearable devices. The paper positions Kolmogorov-Arnold Networks (KANs), which replace fixed activations with learnable spline functions on network edges, as a paradigm shift that delivers parameter efficiency, inherent interpretability, and robustness under data scarcity, enabling transparent clinical EEG systems.

Significance. If the asserted advantages of KANs for EEG tasks are substantiated, the review could help steer the field toward more clinically viable models by emphasizing interpretability and efficiency. The manuscript receives credit for systematically outlining the progression of methods and DL shortcomings. However, as a review without new EEG-specific benchmarks, parameter counts on corpora such as CHB-MIT or TUH, or citations to existing KAN-EEG experiments, its contribution is primarily synthetic rather than demonstrative of the claimed paradigm shift.

major comments (1)
  1. [Abstract] Abstract: The central claim that KANs resolve DL shortcomings by offering 'exceptional parameter efficiency, inherent interpretability for physician trust, and robust performance under data scarcity' for EEG seizure detection is unsupported. The review supplies no EEG dataset results, no direct parameter-count or accuracy comparisons against CNN/RNN baselines, and no citations to prior KAN applications on EEG tasks, rendering the extrapolation from general domains unverified within the manuscript.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback on our review manuscript. We address the single major comment below and will make corresponding revisions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that KANs resolve DL shortcomings by offering 'exceptional parameter efficiency, inherent interpretability for physician trust, and robust performance under data scarcity' for EEG seizure detection is unsupported. The review supplies no EEG dataset results, no direct parameter-count or accuracy comparisons against CNN/RNN baselines, and no citations to prior KAN applications on EEG tasks, rendering the extrapolation from general domains unverified within the manuscript.

    Authors: We agree that the abstract overstates the claims for EEG seizure detection. As a review paper, the manuscript does not contain new experiments, parameter counts on CHB-MIT or TUH, accuracy comparisons, or citations to prior KAN-EEG work. The stated advantages are drawn from the general properties of KANs (Liu et al., 2024) and their results in non-EEG domains. We will revise the abstract to qualify these as potential advantages based on architectural properties, with explicit mention that EEG-specific validation remains future work. We will also add a limitations subsection noting the current lack of EEG benchmarks and the synthetic nature of the review. revision: yes

Circularity Check

0 steps flagged

No circularity: review paper asserts KAN properties via external literature without self-referential derivations

full rationale

The manuscript is a review that catalogs DL limitations for EEG seizure detection and positions KANs as an alternative by describing their general architectural replacement of fixed activations with learnable splines. No equations, new predictions, or fitted parameters are introduced that reduce to the paper's own inputs by construction. Claims about parameter efficiency and interpretability are framed as analysis of cited prior work rather than self-defined or self-cited load-bearing steps. No self-citation chains, ansatzes smuggled via citation, or renamings of known results appear in the provided text. The paper is therefore self-contained as a survey and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The review rests on domain assumptions about DL model limitations and unverified transfer of KAN properties to EEG tasks, drawn from cited literature without new evidence supplied.

axioms (2)
  • domain assumption Standard DL models for EEG operate as opaque black-boxes lacking clinical interpretability and require massive balanced annotated data with high computational costs.
    Stated directly in the abstract as the motivation for exploring alternatives.
  • domain assumption KANs inherently provide parameter efficiency, interpretability, and robustness under data scarcity when applied to EEG seizure detection.
    Central positioning of KANs as the solution without supporting analysis in the abstract.

pith-pipeline@v0.9.1-grok · 5746 in / 1333 out tokens · 18017 ms · 2026-06-26T11:46:23.337470+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

121 extracted references · 6 linked inside Pith

  1. [1]

    Epilepsy

    World Health Organization. Epilepsy. WHO Fact Sheet, February 2024. URL https://www.who.int/ news-room/fact-sheets/detail/epilepsy

    2024

  2. [2]

    Eeg parameters as endpoints in epilepsy clinical trials-an expert panel opinion paper.Epilepsy Research, 187:107028, 2022

    Jeffrey Buchhalter, Caroline Neuray, Jocelyn Y Cheng, O’Neill D’Cruz, Alexandre N Datta, Dennis Dlugos, Jacqueline French, Dietrich Haubenberger, Joseph Hulihan, Pavel Klein, et al. Eeg parameters as endpoints in epilepsy clinical trials-an expert panel opinion paper.Epilepsy Research, 187:107028, 2022

    2022

  3. [3]

    Kan–eeg: towards replacing backbone–mlp for an effective seizure detection system.Royal Society Open Science, 12(3), 2025

    Luis Fernando Herbozo Contreras, Jiashuo Cui, Leping Yu, Zhaojing Huang, Armin Nikpour, and Omid Kavehei. Kan–eeg: towards replacing backbone–mlp for an effective seizure detection system.Royal Society Open Science, 12(3), 2025

    2025

  4. [4]

    Automated real-time detection of tonic-clonic seizures using a wearable emg device.Neurology, 90(5):e428–e434, 2018

    Sándor Beniczky, Isa Conradsen, Oliver Henning, Martin Fabricius, and Peter Wolf. Automated real-time detection of tonic-clonic seizures using a wearable emg device.Neurology, 90(5):e428–e434, 2018

    2018

  5. [5]

    A comprehensive survey on support vector machine classification: Applications, challenges and trends.Neurocomputing, 408: 189–215, 2020

    Jair Cervantes, Farid Garcia-Lamont, Lisbeth Rodríguez-Mazahua, and Asdrubal Lopez. A comprehensive survey on support vector machine classification: Applications, challenges and trends.Neurocomputing, 408: 189–215, 2020

    2020

  6. [6]

    Enhancing k-nearest neighbor algorithm: a comprehensive review and performance analysis of modifications.Journal of Big Data, 11(1):113, 2024

    Rajib Kumar Halder, Mohammed Nasir Uddin, Md Ashraf Uddin, Sunil Aryal, and Ansam Khraisat. Enhancing k-nearest neighbor algorithm: a comprehensive review and performance analysis of modifications.Journal of Big Data, 11(1):113, 2024

    2024

  7. [7]

    A comparative study of decision tree id3 and c4

    Badr Hssina, Abdelkarim Merbouha, Hanane Ezzikouri, and Mohammed Erritali. A comparative study of decision tree id3 and c4. 5.International Journal of Advanced Computer Science and Applications, 4(2):13–19, 2014

    2014

  8. [8]

    Convolutional neural network for detection and classification of seizures in clinical data.Medical & Biological Engineering & Computing, 58(9):1919–1932, 2020

    Tomas Iešmantas and Robertas Alzbutas. Convolutional neural network for detection and classification of seizures in clinical data.Medical & Biological Engineering & Computing, 58(9):1919–1932, 2020

    1919

  9. [9]

    A critical review of recurrent neural networks for sequence learning.arXiv preprint arXiv:1506.00019, 2015

    Zachary C Lipton, John Berkowitz, and Charles Elkan. A critical review of recurrent neural networks for sequence learning.arXiv preprint arXiv:1506.00019, 2015

    Pith/arXiv arXiv 2015

  10. [10]

    Attention is all you need.Advances in neural information processing systems, 30, 2017

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need.Advances in neural information processing systems, 30, 2017

    2017

  11. [11]

    A review of epilepsy detection and prediction methods based on eeg signal processing and deep learning.Frontiers in Neuroscience, 18:1468967, 2024

    Xizhen Zhang, Xiaoli Zhang, Qiong Huang, and Fuming Chen. A review of epilepsy detection and prediction methods based on eeg signal processing and deep learning.Frontiers in Neuroscience, 18:1468967, 2024

    2024

  12. [12]

    Xai4eeg: spectral and spatio-temporal explanation of deep learning-based seizure detection in eeg time series.Neural Computing and Applications, 35(14): 10051–10068, 2023

    Dominik Raab, Andreas Theissler, and Myra Spiliopoulou. Xai4eeg: spectral and spatio-temporal explanation of deep learning-based seizure detection in eeg time series.Neural Computing and Applications, 35(14): 10051–10068, 2023

    2023

  13. [13]

    Deep learning in intracranial eeg for seizure detection: advances, challenges, and clinical applications.Frontiers in Neuroscience, 19:1677898, 2025

    Wasi Ur Rehman Qamar, Min-Ho Lee, and Berdakh Abibullaev. Deep learning in intracranial eeg for seizure detection: advances, challenges, and clinical applications.Frontiers in Neuroscience, 19:1677898, 2025

    2025

  14. [15]

    Graphical insight: revolutionizing seizure detection with eeg representation.Biomedicines, 12(6):1283, 2024

    Muhammad Awais, Samir Brahim Belhaouari, and Khelil Kassoul. Graphical insight: revolutionizing seizure detection with eeg representation.Biomedicines, 12(6):1283, 2024

    2024

  15. [16]

    Epileptic seizure detection using chb-mit dataset: The overlooked perspectives.Royal Society open science, 11(5), 2024

    Emran Ali, Maia Angelova, and Chandan Karmakar. Epileptic seizure detection using chb-mit dataset: The overlooked perspectives.Royal Society open science, 11(5), 2024

    2024

  16. [17]

    Continuous seizure detection based on transformer and long-term ieeg.IEEE Journal of Biomedical and Health Informatics, 26(11):5418–5427, 2022

    Yulin Sun, Weipeng Jin, Xiaopeng Si, Xingjian Zhang, Jiale Cao, Le Wang, Shaoya Yin, and Dong Ming. Continuous seizure detection based on transformer and long-term ieeg.IEEE Journal of Biomedical and Health Informatics, 26(11):5418–5427, 2022

    2022

  17. [18]

    The effectiveness of kolmogorov–arnold networks in the healthcare domain.Applied Sciences, 15(16):9023, 2025

    Vishnu S Pendyala and Nivedita Venkatachalam. The effectiveness of kolmogorov–arnold networks in the healthcare domain.Applied Sciences, 15(16):9023, 2025

    2025

  18. [19]

    Kan: Kolmogorov–arnold networks

    Ziming Liu, Yixuan Wang, Sachin Vaidya, Fabian Ruehle, James Halverson, Marin Soljacic, Thomas Hou, and Max Tegmark. Kan: Kolmogorov–arnold networks. InInternational conference on learning representations, volume 2025, pages 70367–70413, 2025. 25 Evolution of EEG Seizure Detection FrameworksA PREPRINT

    2025

  19. [20]

    Kolmogorov-arnold networks and multi-layer perceptrons: A paradigm shift in neural modeling

    Aradhya Gaonkar, Nihal Jain, Vignesh Chougule, Nikhil Deshpande, Sneha Varur, and Channabasappa Muttal. Kolmogorov-arnold networks and multi-layer perceptrons: A paradigm shift in neural modeling. InInternational Conference on Communication and Computational Technologies, pages 1–15. Springer, 2025

    2025

  20. [21]

    Interpretable clinical classification with kolgomorov-arnold networks.arXiv preprint arXiv:2509.16750, 2025

    Alejandro Almodóvar, Patricia A Apellániz, Alba Garrido, Fernando Fernández-Salvador, Santiago Zazo, and Juan Parras. Interpretable clinical classification with kolgomorov-arnold networks.arXiv preprint arXiv:2509.16750, 2025

    Pith/arXiv arXiv 2025

  21. [22]

    A practitioner’s guide to kolmogorov-arnold networks.arXiv preprint arXiv:2510.25781, 2025

    Amir Noorizadegan, Sifan Wang, Leevan Ling, and Juan P Dominguez-Morales. A practitioner’s guide to kolmogorov-arnold networks.arXiv preprint arXiv:2510.25781, 2025

    Pith/arXiv arXiv 2025

  22. [23]

    Abnormality detection in time-series bio-signals using kolmogorov-arnold networks for resource-constrained devices.MedRxiv, pages 2024–06, 2024

    Zhaojing Huang, Jiashuo Cui, Leping Yu, Luis Fernando Herbozo Contreras, and Omid Kavehei. Abnormality detection in time-series bio-signals using kolmogorov-arnold networks for resource-constrained devices.MedRxiv, pages 2024–06, 2024

    2024

  23. [25]

    On the use of bipolar montages for time-series analysis of intracranial electroencephalograms.Clinical neurophysiology, 117(9):2102–2108, 2006

    Hitten P Zaveri, Robert B Duckrow, and Susan S Spencer. On the use of bipolar montages for time-series analysis of intracranial electroencephalograms.Clinical neurophysiology, 117(9):2102–2108, 2006

    2006

  24. [26]

    The standardized eeg electrode array of the ifcn.Clinical neurophysiology, 128(10): 2070–2077, 2017

    Margitta Seeck, Laurent Koessler, Thomas Bast, Frans Leijten, Christoph Michel, Christoph Baumgartner, Bin He, and Sándor Beniczky. The standardized eeg electrode array of the ifcn.Clinical neurophysiology, 128(10): 2070–2077, 2017

    2070

  25. [27]

    Variability of eeg electrode positions and their underlying brain regions: visualizing gel artifacts from a simultaneous eeg-fmri dataset.Brain and behavior, 12(2):e2476, 2022

    Catriona L Scrivener and Arran T Reader. Variability of eeg electrode positions and their underlying brain regions: visualizing gel artifacts from a simultaneous eeg-fmri dataset.Brain and behavior, 12(2):e2476, 2022

    2022

  26. [28]

    The five percent electrode system for high-resolution eeg and erp measurements.Clinical neurophysiology, 112(4):713–719, 2001

    Robert Oostenveld and Peter Praamstra. The five percent electrode system for high-resolution eeg and erp measurements.Clinical neurophysiology, 112(4):713–719, 2001

    2001

  27. [29]

    10/20, 10/10, and 10/5 systems revisited: their validity as relative head-surface-based positioning systems.Neuroimage, 34(4):1600–1611, 2007

    Valer Jurcak, Daisuke Tsuzuki, and Ippeita Dan. 10/20, 10/10, and 10/5 systems revisited: their validity as relative head-surface-based positioning systems.Neuroimage, 34(4):1600–1611, 2007

    2007

  28. [30]

    Ernst Niedermeyer and F. H. Lopes da Silva.Electroencephalography: Basic Principles, Clinical Applications, and Related Fields. Lippincott Williams & Wilkins, Philadelphia, PA, 2005

    2005

  29. [31]

    Electroencephalography (eeg).The international encyclopedia of communica- tion research methods, pages 1–18, 2017

    Glenna L Read and Isaiah J Innis. Electroencephalography (eeg).The international encyclopedia of communica- tion research methods, pages 1–18, 2017

    2017

  30. [32]

    Quantitative eeg in cognitive neuroscience, 2021

    Yvonne Höller. Quantitative eeg in cognitive neuroscience, 2021

    2021

  31. [33]

    High-frequency oscillations as a new biomarker in epilepsy.Annals of neurology, 71(2):169–178, 2012

    Maeike Zijlmans, Premysl Jiruska, Rina Zelmann, Frans SS Leijten, John GR Jefferys, and Jean Gotman. High-frequency oscillations as a new biomarker in epilepsy.Annals of neurology, 71(2):169–178, 2012

    2012

  32. [34]

    Electroencephalography signal processing: A comprehensive review and analysis of methods and techniques.Sensors, 23(14):6434, 2023

    Ahmad Chaddad, Yihang Wu, Reem Kateb, and Ahmed Bouridane. Electroencephalography signal processing: A comprehensive review and analysis of methods and techniques.Sensors, 23(14):6434, 2023

    2023

  33. [35]

    Physiological artifacts in scalp eeg and ear-eeg.Biomedical engineering online, 16(1):103, 2017

    Simon L Kappel, David Looney, Danilo P Mandic, and Preben Kidmose. Physiological artifacts in scalp eeg and ear-eeg.Biomedical engineering online, 16(1):103, 2017

    2017

  34. [36]

    Eeg datasets for seizure detection and prediction—a review.Epilepsia Open, 8(2):252–267, 2023

    Sheng Wong, Anj Simmons, Jessica Rivera-Villicana, Scott Barnett, Shobi Sivathamboo, Piero Perucca, Zongyuan Ge, Patrick Kwan, Levin Kuhlmann, Rajesh Vasa, et al. Eeg datasets for seizure detection and prediction—a review.Epilepsia Open, 8(2):252–267, 2023

    2023

  35. [37]

    Quantity versus diversity: Influence of data on detecting eeg pathology with advanced ml models.Neural Networks, page 108073, 2025

    Martyna Poziomska, Marian Dovgialo, Przemysław Olbratowski, Paweł Niedbalski, Paweł Ogniewski, Joanna Zych, Jacek Rogala, and Jarosław ˙Zygierewicz. Quantity versus diversity: Influence of data on detecting eeg pathology with advanced ml models.Neural Networks, page 108073, 2025

    2025

  36. [38]

    Long-term ambulatory intracranial eeg.Stereotactic and functional neurosurgery, 103(5):403–414, 2025

    Imran H Quraishi and Lawrence J Hirsch. Long-term ambulatory intracranial eeg.Stereotactic and functional neurosurgery, 103(5):403–414, 2025

    2025

  37. [39]

    The epilepsiae database: An extensive electroencephalography database of epilepsy patients, 2012

    Juliane Klatt, Hinnerk Feldwisch-Drentrup, Matthias Ihle, Vincent Navarro, Markus Neufang, Cesar Teixeira, Claude Adam, Mario Valderrama, Catalina Alvarado-Rojas, Adrien Witon, et al. The epilepsiae database: An extensive electroencephalography database of epilepsy patients, 2012

    2012

  38. [40]

    PhD thesis, Massachusetts Institute of Technology, 2009

    Ali Hossam Shoeb.Application of machine learning to epileptic seizure onset detection and treatment. PhD thesis, Massachusetts Institute of Technology, 2009

    2009

  39. [41]

    Bonn activity maps: Dataset description

    Julian Tanke, Oh-Hun Kwon, Patrick Stotko, Radu Alexandru Rosu, Michael Weinmann, Hassan Errami, Sven Behnke, Maren Bennewitz, Reinhard Klein, Andreas Weber, et al. Bonn activity maps: Dataset description. arXiv preprint arXiv:1912.06354, 2019. 26 Evolution of EEG Seizure Detection FrameworksA PREPRINT

    arXiv 1912

  40. [42]

    Clustering characteristics of uci dataset

    Sun Chang, Yue Shihong, and Li Qi. Clustering characteristics of uci dataset. In2020 39th Chinese control conference (CCC), pages 6301–6306. IEEE, 2020

    2020

  41. [43]

    A dataset of neonatal eeg recordings with seizure annotations.Scientific data, 6(1):190039, 2019

    Nathan J Stevenson, Karoliina Tapani, Leena Lauronen, and Sampsa Vanhatalo. A dataset of neonatal eeg recordings with seizure annotations.Scientific data, 6(1):190039, 2019

    2019

  42. [44]

    The temple university hospital eeg data corpus.Frontiers in neuroscience, 10: 196, 2016

    Iyad Obeid and Joseph Picone. The temple university hospital eeg data corpus.Frontiers in neuroscience, 10: 196, 2016

    2016

  43. [45]

    A classification model of eeg signals based on rnn-lstm for diagnosing focal and generalized epilepsy.Sensors, 22(19):7269, 2022

    Tahereh Najafi, Rosmina Jaafar, Rabani Remli, and Wan Asyraf Wan Zaidi. A classification model of eeg signals based on rnn-lstm for diagnosing focal and generalized epilepsy.Sensors, 22(19):7269, 2022

    2022

  44. [46]

    Combining meta and ensemble learning to classify eeg for seizure detection.Scientific Reports, 15(1):10755, 2025

    Mingze Liu, Jie Liu, Mengna Xu, Yasheng Liu, Jie Li, Weiwei Nie, and Qi Yuan. Combining meta and ensemble learning to classify eeg for seizure detection.Scientific Reports, 15(1):10755, 2025

    2025

  45. [47]

    A review of epileptic seizure detection using machine learning classifiers.Brain informatics, 7(1):5, 2020

    Mohammad Khubeb Siddiqui, Ruben Morales-Menendez, Xiaodi Huang, and Nasir Hussain. A review of epileptic seizure detection using machine learning classifiers.Brain informatics, 7(1):5, 2020

    2020

  46. [48]

    Objective evaluation metrics for automatic classification of eeg events.arXiv preprint arXiv:1712.10107, 2017

    Saeedeh Ziyabari, Vinit Shah, Meysam Golmohammadi, Iyad Obeid, and Joseph Picone. Objective evaluation metrics for automatic classification of eeg events.arXiv preprint arXiv:1712.10107, 2017

    arXiv 2017

  47. [49]

    Jonathan Dan, Una Pale, Alireza Amirshahi, William Cappelletti, Thorir Mar Ingolfsson, Xiaying Wang, Andrea Cossettini, Adriano Bernini, Luca Benini, Sándor Beniczky, et al. Szcore: seizure community open-source research evaluation framework for the validation of electroencephalography-based automated seizure detection algorithms.Epilepsia, 66:14–24, 2025

    2025

  48. [50]

    Removal of artifacts from eeg signals: a review.Sensors, 19(5):987, 2019

    Xiao Jiang, Gui-Bin Bian, and Zean Tian. Removal of artifacts from eeg signals: a review.Sensors, 19(5):987, 2019

    2019

  49. [51]

    Enhancing epilepsy seizure detection through advanced eeg preprocessing techniques and peak-to-peak amplitude fluctuation analysis.Diagnostics, 14(22):2525, 2024

    Muawiyah A Bahhah and Eyad Talal Attar. Enhancing epilepsy seizure detection through advanced eeg preprocessing techniques and peak-to-peak amplitude fluctuation analysis.Diagnostics, 14(22):2525, 2024

    2024

  50. [52]

    Cnn-lstm hybrid deep learning model for remaining useful life estimation.arXiv preprint arXiv:2412.15998, 2024

    Jyosna Philip et al. Cnn-lstm hybrid deep learning model for remaining useful life estimation.arXiv preprint arXiv:2412.15998, 2024

    arXiv 2024

  51. [53]

    The design of digital filters for biomedical signal processing part 3: The design of butterworth and chebychev filters.Journal of biomedical engineering, 5(2):91–102, 1983

    RE Challis and RI Kitney. The design of digital filters for biomedical signal processing part 3: The design of butterworth and chebychev filters.Journal of biomedical engineering, 5(2):91–102, 1983

    1983

  52. [54]

    Scalp eeg classification using deep bi-lstm network for seizure detection.Computers in Biology and Medicine, 124:103919, 2020

    Xinmei Hu, Shasha Yuan, Fangzhou Xu, Yan Leng, Kejiang Yuan, and Qi Yuan. Scalp eeg classification using deep bi-lstm network for seizure detection.Computers in Biology and Medicine, 124:103919, 2020

    2020

  53. [55]

    M Kemal Kıymık,˙Inan Güler, Alper Dizibüyük, and Mehmet Akın. Comparison of stft and wavelet transform methods in determining epileptic seizure activity in eeg signals for real-time application.Computers in biology and medicine, 35(7):603–616, 2005

    2005

  54. [56]

    Automatic seizure detection based on s-transform and deep convolutional neural network.International journal of neural systems, 30(04):1950024, 2020

    Guoyang Liu, Weidong Zhou, and Minxing Geng. Automatic seizure detection based on s-transform and deep convolutional neural network.International journal of neural systems, 30(04):1950024, 2020

    2020

  55. [57]

    Courier Corporation, Mineola, NY , 2004

    Robert Goodell Brown.Smoothing, Forecasting and Prediction of Discrete Time Series. Courier Corporation, Mineola, NY , 2004

    2004

  56. [58]

    Seizure onset zone detection based on convolutional neural networks and eeg signals.Brain Sciences, 14(11):1090, 2024

    Zhejun Kuang, Liming Guo, Jingrui Wang, Jian Zhao, Liu Wang, and Kangwei Geng. Seizure onset zone detection based on convolutional neural networks and eeg signals.Brain Sciences, 14(11):1090, 2024

    2024

  57. [59]

    Two-layer lstm network-based prediction of epileptic seizures using eeg spectral features.Complex & Intelligent Systems, 8(3):2405–2418, 2022

    Kuldeep Singh and Jyoteesh Malhotra. Two-layer lstm network-based prediction of epileptic seizures using eeg spectral features.Complex & Intelligent Systems, 8(3):2405–2418, 2022

    2022

  58. [60]

    Methods of analysis of nonstationary eegs, with emphasis on segmentation techniques: a comparative review.Journal of clinical neurophysiology, 2(3):267–304, 1985

    John S Barlow. Methods of analysis of nonstationary eegs, with emphasis on segmentation techniques: a comparative review.Journal of clinical neurophysiology, 2(3):267–304, 1985

    1985

  59. [61]

    Smote for learning from imbalanced data: progress and challenges, marking the 15-year anniversary.Journal of artificial intelligence research, 61:863–905, 2018

    Alberto Fernández, Salvador Garcia, Francisco Herrera, and Nitesh V Chawla. Smote for learning from imbalanced data: progress and challenges, marking the 15-year anniversary.Journal of artificial intelligence research, 61:863–905, 2018

    2018

  60. [62]

    Knnor: An oversampling technique for imbalanced datasets.Applied soft computing, 115:108288, 2022

    Ashhadul Islam, Samir Brahim Belhaouari, Atiq Ur Rehman, and Halima Bensmail. Knnor: An oversampling technique for imbalanced datasets.Applied soft computing, 115:108288, 2022

    2022

  61. [63]

    Applications of artificial intelligence in epilepsy.International Journal of Advanced Medical and Health Research, 8(2):41–48, 2021

    Pradeep Pankajakshan Nair, Rajeswari Aghoram, and Madhuri Laxman Khilari. Applications of artificial intelligence in epilepsy.International Journal of Advanced Medical and Health Research, 8(2):41–48, 2021

    2021

  62. [64]

    Extracting and selecting distinctive eeg features for efficient epileptic seizure prediction.IEEE journal of biomedical and health informatics, 19(5):1648–1659, 2014

    Ning Wang and Michael R Lyu. Extracting and selecting distinctive eeg features for efficient epileptic seizure prediction.IEEE journal of biomedical and health informatics, 19(5):1648–1659, 2014. 27 Evolution of EEG Seizure Detection FrameworksA PREPRINT

    2014

  63. [65]

    Eeg signals feature extraction based on dwt and emd combined with approximate entropy.Brain sciences, 9(8):201, 2019

    Na Ji, Liang Ma, Hui Dong, and Xuejun Zhang. Eeg signals feature extraction based on dwt and emd combined with approximate entropy.Brain sciences, 9(8):201, 2019

    2019

  64. [66]

    Current trends in feature extraction and classification method- ologies of biomedical signals.Current Medical Imaging, 20(1):E090323214502, 2024

    Sachin Kumar, Karan Veer, and Sanjeev Kumar. Current trends in feature extraction and classification method- ologies of biomedical signals.Current Medical Imaging, 20(1):E090323214502, 2024

    2024

  65. [67]

    Time–frequency feature representation using energy concentration: An overview of recent advances.Digital signal processing, 19(1):153–183, 2009

    Ervin Sejdi´c, Igor Djurovi´c, and Jin Jiang. Time–frequency feature representation using energy concentration: An overview of recent advances.Digital signal processing, 19(1):153–183, 2009

    2009

  66. [68]

    Statistical graph signal processing: Stationarity and spectral estimation.Cooperative and Graph Signal Processing, pages 325–347, 2018

    Santiago Segarra, Sundeep Prabhakar Chepuri, Antonio G Marques, and Geert Leus. Statistical graph signal processing: Stationarity and spectral estimation.Cooperative and Graph Signal Processing, pages 325–347, 2018

    2018

  67. [69]

    Wind speed time series synthesis using a parametrized power spectral density function.Wind Energy Science Discussions, 2023:1–21, 2023

    Ram C Poudel, David Corbus, and Ian Baring-Gould. Wind speed time series synthesis using a parametrized power spectral density function.Wind Energy Science Discussions, 2023:1–21, 2023

    2023

  68. [70]

    Eeg signal analysis: a survey

    D Puthankattil Subha, Paul K Joseph, Rajendra Acharya U, and Choo Min Lim. Eeg signal analysis: a survey. Journal of medical systems, 34(2):195–212, 2010

    2010

  69. [71]

    Epileptic seizure detection using machine learning: A systematic review and meta-analysis.Brain Sciences, 15(6):634, 2025

    Lin Bai, Gerhard Litscher, and Xiaoning Li. Epileptic seizure detection using machine learning: A systematic review and meta-analysis.Brain Sciences, 15(6):634, 2025

    2025

  70. [72]

    A review of feature selection methods for machine learning-based disease risk prediction.Frontiers in bioinformatics, 2:927312, 2022

    Nicholas Pudjihartono, Tayaza Fadason, Andreas W Kempa-Liehr, and Justin M O’Sullivan. A review of feature selection methods for machine learning-based disease risk prediction.Frontiers in bioinformatics, 2:927312, 2022

    2022

  71. [73]

    A tutorial on principal component analysis.arXiv preprint arXiv:1404.1100, 2014

    Jonathon Shlens. A tutorial on principal component analysis.arXiv preprint arXiv:1404.1100, 2014

    Pith/arXiv arXiv 2014

  72. [74]

    A comprehensive review on discriminant analysis for addressing challenges of class-level limitations, small sample size, and robustness.Processes, 12(7):1382, 2024

    Lingxiao Qu and Yan Pei. A comprehensive review on discriminant analysis for addressing challenges of class-level limitations, small sample size, and robustness.Processes, 12(7):1382, 2024

    2024

  73. [75]

    Umap: Uniform manifold approximation and projection for dimension reduction.arXiv preprint arXiv:1802.03426, 2018

    Leland McInnes, John Healy, and James Melville. Umap: Uniform manifold approximation and projection for dimension reduction.arXiv preprint arXiv:1802.03426, 2018

    Pith/arXiv arXiv 2018

  74. [76]

    An automated data mining framework using autoencoders for feature extraction and dimensionality reduction

    Yaxin Liang, Xinshi Li, Xin Huang, Ziqi Zhang, and Yue Yao. An automated data mining framework using autoencoders for feature extraction and dimensionality reduction. In2024 4th International Conference on Electronic Information Engineering and Computer Communication (EIECC), pages 710–714. IEEE, 2024

    2024

  75. [77]

    Theoretical foundations of t-sne for visualizing high-dimensional clustered data

    T Tony Cai and Rong Ma. Theoretical foundations of t-sne for visualizing high-dimensional clustered data. Journal of Machine Learning Research, 23(301):1–54, 2022

    2022

  76. [78]

    A survey: Potential dimensionality reduction methods.arXiv preprint arXiv:2502.11036, 2025

    Yuan-chin Ivan Chang. A survey: Potential dimensionality reduction methods.arXiv preprint arXiv:2502.11036, 2025

    arXiv 2025

  77. [79]

    A study on dimen- sionality reduction and parameters for hyperspectral imagery based on manifold learning.Sensors, 24(7):2089, 2024

    Wenhui Song, Xin Zhang, Guozhu Yang, Yijin Chen, Lianchao Wang, and Hanghang Xu. A study on dimen- sionality reduction and parameters for hyperspectral imagery based on manifold learning.Sensors, 24(7):2089, 2024

    2089

  78. [80]

    Logistic regression in data analysis: an overview.International Journal of Data Analysis Techniques and Strategies, 3(3):281–299, 2011

    Maher Maalouf. Logistic regression in data analysis: an overview.International Journal of Data Analysis Techniques and Strategies, 3(3):281–299, 2011

    2011

  79. [81]

    Random forests.Machine learning, 45(1):5–32, 2001

    Leo Breiman. Random forests.Machine learning, 45(1):5–32, 2001

    2001

  80. [82]

    Gradient boosting machine: a survey.arXiv preprint arXiv:1908.06951, 2019

    Zhiyuan He, Danchen Lin, Thomas Lau, and Mike Wu. Gradient boosting machine: a survey.arXiv preprint arXiv:1908.06951, 2019

    arXiv 1908

Showing first 80 references.