Generalizable Sleep Staging via Multi-Level Domain Alignment
This work addresses the challenge of generalizing sleep staging models to unseen datasets, which is crucial for broader clinical application, though it is incremental as it builds on existing domain generalization methods.
The paper tackles the problem of limited generalization in automatic sleep staging by introducing domain generalization and proposing the SleepDG framework, which achieves state-of-the-art performance on five public datasets.
Automatic sleep staging is essential for sleep assessment and disorder diagnosis. Most existing methods depend on one specific dataset and are limited to be generalized to other unseen datasets, for which the training data and testing data are from the same dataset. In this paper, we introduce domain generalization into automatic sleep staging and propose the task of generalizable sleep staging which aims to improve the model generalization ability to unseen datasets. Inspired by existing domain generalization methods, we adopt the feature alignment idea and propose a framework called SleepDG to solve it. Considering both of local salient features and sequential features are important for sleep staging, we propose a Multi-level Feature Alignment combining epoch-level and sequence-level feature alignment to learn domain-invariant feature representations. Specifically, we design an Epoch-level Feature Alignment to align the feature distribution of each single sleep epoch among different domains, and a Sequence-level Feature Alignment to minimize the discrepancy of sequential features among different domains. SleepDG is validated on five public datasets, achieving the state-of-the-art performance.