Toward Foundational Model for Sleep Analysis Using a Multimodal Hybrid Self-Supervised Learning Framework

TL;DR

SynthSleepNet is a multimodal self-supervised learning framework that significantly improves sleep analysis accuracy and robustness across multiple tasks, even with limited labeled data, by leveraging diverse physiological signals and advanced contextual modeling.

Contribution

The paper introduces SynthSleepNet, a novel hybrid self-supervised framework that effectively integrates multimodal physiological signals for sleep analysis, outperforming existing methods.

Findings

Achieved 89.89% sleep-stage classification accuracy

Reached 99.75% apnea detection accuracy

Maintained high performance with limited labels in semi-supervised settings

Abstract

Sleep is essential for maintaining human health and quality of life. Analyzing physiological signals during sleep is critical in assessing sleep quality and diagnosing sleep disorders. However, manual diagnoses by clinicians are time-intensive and subjective. Despite advances in deep learning that have enhanced automation, these approaches remain heavily dependent on large-scale labeled datasets. This study introduces SynthSleepNet, a multimodal hybrid self-supervised learning framework designed for analyzing polysomnography (PSG) data. SynthSleepNet effectively integrates masked prediction and contrastive learning to leverage complementary features across multiple modalities, including electroencephalogram (EEG), electrooculography (EOG), electromyography (EMG), and electrocardiogram (ECG). This approach enables the model to learn highly expressive representations of PSG data. Furthermore, a temporal context module based on Mamba was developed to efficiently capture contextual information across signals. SynthSleepNet achieved superior performance compared to state-of-the-art methods across three downstream tasks: sleep-stage classification, apnea detection, and hypopnea detection, with accuracies of 89.89%, 99.75%, and 89.60%, respectively. The model demonstrated robust performance in a semi-supervised learning environment with limited labels, achieving accuracies of 87.98%, 99.37%, and 77.52% in the same tasks. These results underscore the potential of the model as a foundational tool for the comprehensive analysis of PSG data. SynthSleepNet demonstrates comprehensively superior performance across multiple downstream tasks compared to other methodologies, making it expected to set a new standard for sleep disorder monitoring and diagnostic systems.