Toward Foundational Model for Sleep Analysis Using a Multimodal Hybrid-Self-Supervised Learning Framework
Open Access
- 9 September 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 55 (12) , 5619-5632
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 (SSL) 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 TCM 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-SSL 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. The source code is available at https://github.com/dlcjfgmlnasa/SynthSleepNet.Keywords
All Related Versions
Funding Information
- National Research Foundation of Korea (NRF) grant funded by the Korea government (2022R1A2C1013205)
This publication has 35 references indexed in Scilit:
- RAFNet: Restricted attention fusion network for sleep apnea detectionNeural Networks, 2023
- MultiMAE: Multi-modal Multi-task Masked Autoencoders2022
- SleepExpertNet: high-performance and class-balanced deep learning approach inspired from the expert neurologists for sleep stage classificationJournal of Ambient Intelligence and Humanized Computing, 2022
- mulEEG: A Multi-view Representation Learning on EEG Signals2022
- Intra- and inter-epoch temporal context network (IITNet) using sub-epoch features for automatic sleep scoring on raw single-channel EEGBiomedical Signal Processing and Control, 2020
- Neurological Deficits in Obstructive Sleep ApneaCurrent Treatment Options in Neurology, 2017
- Automatic detection of spindles and K-complexes in sleep EEG using switching multiple modelsBiomedical Signal Processing and Control, 2014
- Chapter 3 An overview of polysomnography2005
- The Effects of Age, Sex, Ethnicity, and Sleep-Disordered Breathing on Sleep ArchitectureArchives of internal medicine (1960), 2004
- Obstructive sleep apnoeaThe Lancet, 2002