EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Spiking neural networks with liquid time-constant dynamics and dendritic branches for efficient time-domain edge-based seizure detection

Luis Fernando Herbozo Contreras, Leping Yu, Zhaojing Huang, Isabelle Aguilar, Armin Nikpour, Omid Kavehei

Machine Learning: Health · 2025

DOI: 10.1088/3049-477x/adeb8e ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machine Learning: Health, 2025. Verfasst von Luis Fernando Herbozo Contreras und 5 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Australian Research Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Epilepsy is a significant global health issue, requiring dependable diagnostic tools like scalp encephalogram (scalp-EEG), subscalp EEG, and intracranial EEG for precise seizure detection and treatment. artificial intelligence (AI) has emerged as a powerful tool in this domain, offering the potential for real-time, responsive monitoring. Traditional methods often rely on feature extraction techniques like Short-Time Fourier Transform (STFT), which can increase power consumption, making them less suitable for deployment on edge devices. While large models can improve accuracy without STFT, their size also limits their practicality for edge applications. This study introduces liquid-dendrite, a novel bio-inspired model for seizure detection, leveraging liquid-time constant spiking neurons and dendrites spiking neurons with heterogeneous time-constants. The model comprises two hidden layers with dendritic neurons and one layer of liquid-time constant networks. Our model achieves a memory efficacy of 535 KB with 130 K trainable parameters. The model was tested across the most noteworthy epilepsy datasets for scalp EEG (TUH and CHB-MIT) and iEEG (EPILEPSIAE). Our model demonstrated commendable performance, achieving area under the receiver operating characteristic scores of 83%, 96%, and 93%, respectively, outperforming some existing models in an energy and memory-efficient way. Moreover, we conducted a robustness test by blacking out EEG channels at the inference stage, where we showed the ability of our network to work with fewer channels. We deployed our tiny model and perform inference at the edge of the Raspberry Pi 5 without the need for additional quantization. This highlights the potential of Neuro-Inspired AI for efficient, small-scale, and energy-embedded AI systems across different brain modalities.

Förderung

  • Australian Research Council

Zitieren

APA

Contreras, L. F. H., Yu, L., Huang, Z., Aguilar, I., Nikpour, A., & Kavehei, O. (2025). Spiking neural networks with liquid time-constant dynamics and dendritic branches for efficient time-domain edge-based seizure detection. Machine Learning: Health. https://doi.org/10.1088/3049-477x/adeb8e

Vancouver

Contreras LFH, Yu L, Huang Z, Aguilar I, Nikpour A, Kavehei O. Spiking neural networks with liquid time-constant dynamics and dendritic branches for efficient time-domain edge-based seizure detection. Machine Learning: Health. 2025. doi:10.1088/3049-477x/adeb8e

BibTeX
@article{Contreras2025,
  author = {Contreras, Luis Fernando Herbozo and Yu, Leping and Huang, Zhaojing and Aguilar, Isabelle and Nikpour, Armin and Kavehei, Omid},
  title = {Spiking neural networks with liquid time-constant dynamics and dendritic branches for efficient time-domain edge-based seizure detection},
  journal = {Machine Learning: Health},
  year = {2025},
  doi = {10.1088/3049-477x/adeb8e}
}
Kostenlos starten →