SleepBand: Single-Source Domain Generalization for Sleep Staging via Physiologically Structured Spectral Modeling
It addresses the practical problem of generalizing sleep staging models to unseen datasets without requiring multiple source domains or domain labels, which is a common limitation in clinical applications.
SleepBand tackles single-source domain generalization for sleep staging by embedding oscillatory priors via a learnable Morlet filter bank, achieving state-of-the-art performance on five public datasets and remaining competitive under multi-source DG.
Generalizing sleep staging models to unseen datasets is challenging, and typical domain generalization (DG) methods often rely on multiple source domains or domain labels that are rarely available in practice. We tackle the stricter and more practical setting of single-source domain generalization: training on a single labeled source dataset, without domain labels or access to target data. We present SleepBand, a physiology-guided framework that embeds oscillatory priors via a learnable Morlet filter bank and a structured integration-and-recalibration pipeline. This anchors representations to domain-invariant sleep rhythms (e.g., slow waves, spindles), reducing reliance on dataset-specific artefacts. On five public datasets, SleepBand achieves state-of-the-art SDG performance and remains competitive under leave-one-domain-out (multi-source) DG. Analyses show that the learned filters align with canonical neurophysiology and that robustness stems from focusing on narrowband, physiologically meaningful cues. Our results suggest that principled, physiology-aware inductive biases are a promising path for robust single-domain sleep staging. Code is available at https://github.com/lzcn/sleep-band