S4Sleep: Elucidating the design space of deep-learning-based sleep stage classification models

TL;DR

This paper systematically explores architectural choices in deep learning models for sleep stage classification, identifying robust encoder-predictor architectures with structured state space models that outperform existing methods on a large dataset.

Contribution

It provides a comprehensive analysis of design decisions in encoder-predictor architectures, introducing effective configurations with structured state space models for sleep staging.

Findings

Identified robust architectures applicable to time series and spectrogram inputs.

Achieved statistically significant performance improvements over state-of-the-art methods.

Demonstrated the generalizability of the architecture search methodology.

Abstract

Scoring sleep stages in polysomnography recordings is a time-consuming task plagued by significant inter-rater variability. Therefore, it stands to benefit from the application of machine learning algorithms. While many algorithms have been proposed for this purpose, certain critical architectural decisions have not received systematic exploration. In this study, we meticulously investigate these design choices within the broad category of encoder-predictor architectures. We identify robust architectures applicable to both time series and spectrogram input representations. These architectures incorporate structured state space models as integral components and achieve statistically significant performance improvements compared to state-of-the-art approaches on the extensive Sleep Heart Health Study dataset. We anticipate that the architectural insights gained from this study along with the refined methodology for architecture search demonstrated herein will not only prove valuable for future research in sleep staging but also hold relevance for other time series annotation tasks.