Multi-Scale Residual Graph Learning with Contextual Memory for Sleep Staging

摘要

Sleep is a vital physiological process, and its quality directly impacts human health. Although deep learning-based automated sleep staging methods have achieved significant progress, the studies still face limitations in fully mining spatio-temporal graph dependencies. Existing methods face challenges in capturing implicit spatial topology among multi-channel signals and efficiently fusing multi-level features in long-sequence modeling. To address these challenges, this paper proposes a novel hybrid deep learning framework for sleep staging that progressively and explicitly integrates spatial topology modeling, cross-level feature fusion, and global temporal aggregation. By decoupling shallow and deep spatial features and hierarchically organizing feature interactions, the proposed framework effectively captures implicit spatial dependencies while maintaining computational efficiency in long-sequence modeling. Extensive experimental results on multiple sleep staging datasets (83.1% accuracy, 80.1% F1-score on ISRUC-S1, 84.0% accuracy, 83.8% F1-score on ISRUC-S3, and 85.9% accuracy, 83.7% F1-score on Sleep-EDF-153) demonstrate that the proposed method achieves competitive accuracy with significantly reduced computational latency, highlighting its strong potential for mobile and real-time sleep monitoring applications.