EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Lightweight Neural Network Based on Memory and Transition Probability for Accurate Real-Time Sleep Stage Classification

Dhanushka Wijesinghe iD, Ivan T Lima iD

Brain Sciences · 25.07.2025 · 15 (8) · 789

DOI: 10.3390/brainsci15080789 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Brain Sciences, online erschienen am 25.07.2025. Verfasst von Dhanushka Wijesinghe und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical and Computer Engineering, North Dakota State University, Fargo, ND 58105, USA. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NDSU Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: This study shows a lightweight hybrid framework based on a feedforward neural network using a single frontopolar electroencephalography channel, which is a practical configuration for wearable systems combining memory and a sleep stage transition probability matrix.

Methods

Motivated by autocorrelation analysis, revealing strong temporal dependencies across sleep stages, we incorporate prior epoch information as additional features. To capture temporal context without requiring long input sequences, we introduce a transition-aware feature derived from the softmax output of the previous epoch, weighted by a learned stage transition matrix. The model combines predictions from memory-based and no-memory networks using a confidence-driven fallback strategy.

Results

The proposed model achieves up to 85.4% accuracy and 0.79 Cohen’s kappa, despite using only a single 30 s epoch per prediction. Compared to other models that use a single frontopolar channel, our method outperforms convolutional neural networks, recurrent neural networks, and decision tree approaches. Additionally, confidence-based rejection of low-certainty predictions enhances reliability, since most of the epochs with low confidence in the sleep stage classification contain transitions between sleep stages.

Conclusions

These results demonstrate that the proposed method balances performance, interpretability, and computational efficiency, making it well-suited for real-time clinical and wearable sleep staging applications using battery-powered computing devices.

Schlagwörter

Stichwörter der Autor:innen

EEGautocovariancefeedforward neural networkmemory-augmented neural networkportable sleep monitoringreal-time classification

Autor:innen und Institutionen

  • Department of Electrical and Computer Engineering, North Dakota State University, Fargo, ND 58105, USA.

ORCID-Profile: Dhanushka Wijesinghe, Ivan T Lima

Förderung

  • NDSU Foundation

Zitieren

APA

Wijesinghe, D., & Lima, I. T. (2025). A Lightweight Neural Network Based on Memory and Transition Probability for Accurate Real-Time Sleep Stage Classification. Brain Sciences, 15(8), 789. https://doi.org/10.3390/brainsci15080789

Vancouver

Wijesinghe D, Lima IT. A Lightweight Neural Network Based on Memory and Transition Probability for Accurate Real-Time Sleep Stage Classification. Brain Sciences. 2025;15(8):789. doi:10.3390/brainsci15080789

BibTeX
@article{Wijesinghe2025,
  author = {Wijesinghe, Dhanushka and Lima, Ivan T},
  title = {A Lightweight Neural Network Based on Memory and Transition Probability for Accurate Real-Time Sleep Stage Classification},
  journal = {Brain Sciences},
  year = {2025},
  volume = {15},
  number = {8},
  pages = {789},
  doi = {10.3390/brainsci15080789}
}
Kostenlos starten →