A Systematic Evaluation of Self-Supervised Learning for Label-Efficient Sleep Staging with Wearable EEG

2025-10-09

AI Summary

To address the scarcity of labeled data and high annotation costs for wearable EEG-based sleep staging, this work presents the first systematic investigation of self-supervised learning (SSL) in this domain. We propose three evaluation paradigms and conduct comprehensive comparisons of leading SSL methods on two real-world datasets—BOAS and HOGAR. Results demonstrate that SSL models achieve >80% accuracy using only 5–10% labeled data, outperforming fully supervised baselines by approximately 10%. Crucially, these models exhibit strong generalization across diverse populations, recording environments, and low signal-to-noise ratio conditions. This study not only validates the efficacy of SSL for resource-constrained neurophysiological signal analysis but also establishes a low-labeling benchmark framework for wearable EEG sleep staging—substantially lowering barriers to clinical deployment.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Wearable EEG devices have emerged as a promising alternative to polysomnography (PSG). As affordable and scalable solutions, their widespread adoption results in the collection of massive volumes of unlabeled data that cannot be analyzed by clinicians at scale. Meanwhile, the recent success of deep learning for sleep scoring has relied on large annotated datasets. Self-supervised learning (SSL) offers an opportunity to bridge this gap, leveraging unlabeled signals to address label scarcity and reduce annotation effort. In this paper, we present the first systematic evaluation of SSL for sleep staging using wearable EEG. We investigate a range of well-established SSL methods and evaluate them on two sleep databases acquired with the Ikon Sleep wearable EEG headband: BOAS, a high-quality benchmark containing PSG and wearable EEG recordings with consensus labels, and HOGAR, a large collection of home-based, self-recorded, and unlabeled recordings. Three evaluation scenarios are defined to study label efficiency, representation quality, and cross-dataset generalization. Results show that SSL consistently improves classification performance by up to 10% over supervised baselines, with gains particularly evident when labeled data is scarce. SSL achieves clinical-grade accuracy above 80% leveraging only 5% to 10% of labeled data, while the supervised approach requires twice the labels. Additionally, SSL representations prove robust to variations in population characteristics, recording environments, and signal quality. Our findings demonstrate the potential of SSL to enable label-efficient sleep staging with wearable EEG, reducing reliance on manual annotations and advancing the development of affordable sleep monitoring systems.

Problem

• Evaluating self-supervised learning for label-efficient wearable EEG sleep staging
• Addressing label scarcity in sleep staging using unlabeled wearable EEG data
• Improving classification performance with limited labeled data through SSL methods

Innovation

• Self-supervised learning for wearable EEG sleep staging
• SSL improves performance with limited labeled data
• SSL achieves clinical accuracy using minimal annotations