<ul><li>Accurate classification of sleep stages based on bio-signals is essential for automated sleep stage annotation.</li><li>Deep learning methods have shown promise in automating the sleep stage classification task, but face challenges such as the need for large labeled datasets, inter-individual variability, and overlooking high-order relationships among bio-signals.</li><li>To overcome these challenges, MetaSTH-Sleep is proposed, a few-shot sleep stage classification framework using spatial-temporal hypergraph enhanced meta-learning.</li><li>Experimental results show that MetaSTH-Sleep significantly improves performance across various subjects, providing valuable support for clinicians in sleep stage annotation.</li></ul>

MetaSTH-Sleep: Towards Effective Few-Shot Sleep Stage Classification with Spatial-Temporal Hypergraph Enhanced Meta-Learning

Discover more