EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MMFNet: A multi-branch multi-scale framework with adaptive sparse self-attention and cross-modal fusion for sleep stage assessment

Yuan Li, Ningning Wang iD

PLOS One · 17.07.2026 · 21 (7) · e0353930

DOI: 10.1371/journal.pone.0353930 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 17.07.2026. Verfasst von Yuan Li und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Physics and Electronic-Electrical Engineering, ABA Teachers College, Aba Tibetan and Qiang Autonomous Prefecture, Sichuan, China. Hauptthemen laut MEDLINE: Sleep Stages. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Aba Teachers College Scientific Research Funding Projects.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate sleep stage classification serves as a crucial foundation for sleep health assessment and disease diagnosis. However, existing approaches still encounter several challenges, including limited feature representation, redundancy in extracted information, and difficulties in effectively integrating cross-modal physiological signals. To address these issues, we propose a novel framework entitled Multi-Branch Multi-Scale Fusion with Adaptive Sparse Self-Attention and Cross-Modal Integration (MFFNet). Specifically, considering the prominent time-frequency characteristics of sleep signals, multiple branches are designed to capture multi-scale information, including a time-granular feature extraction branch and a time-frequency extraction branch. Furthermore, a multi-head adaptive sparse self-attention mechanism is introduced to suppress redundant information while emphasizing discriminative features. In addition, we employ an adaptive cross-modal fusion strategy that dynamically integrates information from EEG and EOG, and further visualize the contribution of each modality to sleep stage classification. Experiments conducted on the Sleep-EDF-39 and Sleep-EDF-153 datasets demonstrate the effectiveness of the proposed approach. Using the Fpz-Cz EEG channel and EOG signals, MFFNet achieves accuracy rates of 84.26% and 81.86%, with F1 scores of 75.91% and 73.38%, respectively, highlighting its competitive performance in sleep stage assessment.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Sleep StagesHumansElectroencephalographyAlgorithmsElectrooculographySignal Processing, Computer-AssistedPolysomnography

Autor:innen und Institutionen

  • School of Physics and Electronic-Electrical Engineering, ABA Teachers College, Aba Tibetan and Qiang Autonomous Prefecture, Sichuan, China.

ORCID-Profile: Ningning Wang

Förderung

  • Aba Teachers College Scientific Research Funding Projects

Zitieren

APA

Li, Y., & Wang, N. (2026). MMFNet: A multi-branch multi-scale framework with adaptive sparse self-attention and cross-modal fusion for sleep stage assessment. PLOS One, 21(7), e0353930. https://doi.org/10.1371/journal.pone.0353930

Vancouver

Li Y, Wang N. MMFNet: A multi-branch multi-scale framework with adaptive sparse self-attention and cross-modal fusion for sleep stage assessment. PLOS One. 2026;21(7):e0353930. doi:10.1371/journal.pone.0353930

BibTeX
@article{Li2026,
  author = {Li, Yuan and Wang, Ningning},
  title = {MMFNet: A multi-branch multi-scale framework with adaptive sparse self-attention and cross-modal fusion for sleep stage assessment},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {7},
  pages = {e0353930},
  doi = {10.1371/journal.pone.0353930}
}
Kostenlos starten →