EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Towards Interpretable Seizure Detection: An Excitation/Inhibition Dynamic Polynomial Network Framework for Electroencephalography

Xihan Sun, Ying Yan iD, Na Liu, Shencun Fang, Jun Cai iD, Edmond Qi Wu, Aiguo Song iD, Junjie Xu

Sensors · 01.06.2026 · 26 (11) · 3488

DOI: 10.3390/s26113488 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 01.06.2026. Verfasst von Xihan Sun und 7 weiteren Autor:innen. Beteiligte Einrichtung: Reading Academy, Nanjing University of Information Science and Technology, Nanjing 210044, China. Hauptthemen laut MEDLINE: Seizures, Electroencephalography. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, the Natural Science Foundation of Jiangsu Province, the Open Foundation of the Key Laboratory of Technology and Equipment of Tianjin Urban Air Transportation System.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Epilepsy is a prevalent neurological disorder characterized by recurrent seizures, and electroencephalogram (EEG) signals provide a direct measure of brain activity for detection. Although deep learning achieves high accuracy, it often lacks physiological interpretability. We propose the Excitation/Inhibition Dynamic Polynomial Network (E/I-DynPolyNet), a biologically grounded framework for interpretable seizure detection. Specifically, E/I-DynPolyNet introduces a dual excitatory/inhibitory (E/I) pathway with sign-constrained synaptic weights, encouraging the learned activations to reflect latent E/I representations. Furthermore, a differentiable Wilson-Cowan (WC) module is embedded to govern the temporal evolution of E/I interactions, ensuring consistency with neurophysiological principles. A physics-informed optimization strategy integrates supervised learning with dynamical residual constraints and E/I balance regularization, guiding the model to learn physiologically consistent representations. Experimental results on the CHB-MIT and Bonn datasets demonstrate competitive accuracies of 95.81% and 98.5%, respectively. Crucially, E/I-DynPolyNet enables quantitative estimation of E/I imbalance, revealing that E/I ratios increase from 1.01 in the pre-ictal phase to 1.38 during seizures—a finding consistent with clinical observations of ictogenesis. These results indicate that E/I-DynPolyNet not only improves detection performance but also provides a mechanistic description of seizure dynamics, bridging the gap between data-driven learning and neurophysiological interpretation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SeizuresElectroencephalographyHumansAlgorithmsSignal Processing, Computer-AssistedNeural Networks, Computer

Stichwörter der Autor:innen

Wilson–Cowan dynamicsexcitatory/inhibitory imbalancephysics-informed neural networksseizure detection

Autor:innen und Institutionen

  • Reading Academy, Nanjing University of Information Science and Technology, Nanjing 210044, China.
  • School of Automation, Jiangsu Engineering Research Center on Meteorological Energy Using and Control (C-MEIC), Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology (CICAEET), and S…
  • School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei 230009, China.
  • Department of Respiratory Medicine, The Affiliated Brain Hospital of Nanjing Medical University, Nanjing 210029, China.
  • Department of Automation, Shanghai Jiao Tong University, Shanghai 200240, China.
  • School of Instrument Science and Engineering, Southeast University, Nanjing 210096, China.

ORCID-Profile: Ying Yan, Jun Cai, Aiguo Song

Förderung

  • National Natural Science Foundation of China
  • the Natural Science Foundation of Jiangsu Province
  • the Open Foundation of the Key Laboratory of Technology and Equipment of Tianjin Urban Air Transportation System
  • the Excellent Research and Innovation Team Project of Universities in Anhui Province
  • the Research on quality Assurance and Evaluation of higher Education in Jiangsu Province
  • Natural Science Foundation of Jiangsu Province
  • Open Foundation of the Key Laboratory of Technology and Equipment of Tianjin Urban Air Transportation System
  • Excellent Research and Innovation Team Project of Universities in Anhui Province

Zitieren

APA

Sun, X., Yan, Y., Liu, N., Fang, S., Cai, J., Wu, E. Q., Song, A., & Xu, J. (2026). Towards Interpretable Seizure Detection: An Excitation/Inhibition Dynamic Polynomial Network Framework for Electroencephalography. Sensors, 26(11), 3488. https://doi.org/10.3390/s26113488

Vancouver

Sun X, Yan Y, Liu N, Fang S, Cai J, Wu EQ, et al. Towards Interpretable Seizure Detection: An Excitation/Inhibition Dynamic Polynomial Network Framework for Electroencephalography. Sensors. 2026;26(11):3488. doi:10.3390/s26113488

BibTeX
@article{Sun2026,
  author = {Sun, Xihan and Yan, Ying and Liu, Na and Fang, Shencun and Cai, Jun and Wu, Edmond Qi and Song, Aiguo and Xu, Junjie},
  title = {Towards Interpretable Seizure Detection: An Excitation/Inhibition Dynamic Polynomial Network Framework for Electroencephalography},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {11},
  pages = {3488},
  doi = {10.3390/s26113488}
}
Kostenlos starten →