Combining scEEG and PPG for reliable sleep staging using lightweight wearables
arXiv:2602.15042v1 Announce Type: cross Abstract: Reliable sleep staging remains challenging for lightweight wearable devices such as single-channel electroencephalography (scEEG) or photoplethysmography (PPG). scEEG offers direct measurement of cortical activity and serves as the foundation for sleep staging, yet exhibits limited performance on light sleep stages. PPG provides a low-cost complement that captures autonomic signatures effective for detecting light sleep. However, prior PPG-based methods rely on full night recordings (8 - 10 hours) as input context, which is less practical to provide timely feedback for sleep intervention. In this work, we investigate scEEG-PPG fusion for 4-class sleep staging under short-window (30 s - 30 min) constraints. First, we evaluate the temporal context required for each modality, to better understand the relationship of sleep staging performance with respect to monitoring window. Second, we investigate three fusion strategies: score-level fus
arXiv:2602.15042v1 Announce Type: cross Abstract: Reliable sleep staging remains challenging for lightweight wearable devices such as single-channel electroencephalography (scEEG) or photoplethysmography (PPG). scEEG offers direct measurement of cortical activity and serves as the foundation for sleep staging, yet exhibits limited performance on light sleep stages. PPG provides a low-cost complement that captures autonomic signatures effective for detecting light sleep. However, prior PPG-based methods rely on full night recordings (8 - 10 hours) as input context, which is less practical to provide timely feedback for sleep intervention. In this work, we investigate scEEG-PPG fusion for 4-class sleep staging under short-window (30 s - 30 min) constraints. First, we evaluate the temporal context required for each modality, to better understand the relationship of sleep staging performance with respect to monitoring window. Second, we investigate three fusion strategies: score-level fusion, cross-attention fusion enabling feature-level interactions, and Mamba-enhanced fusion incorporating temporal context modeling. Third, we train and evaluate on the Multi-Ethnic Study of Atherosclerosis (MESA) dataset and perform cross-dataset validation on the Cleveland Family Study (CFS) and the Apnea, Bariatric surgery, and CPAP (ABC) datasets. The Mamba-enhanced fusion achieves the best performance on MESA (Cohen's Kappa $\kappa$ = 0.798, Acc = 86.9%), with particularly notable improvement in light sleep classification (F1-score: 85.63% vs. 77.76%, recall: 82.85% vs. 69.95% for scEEG alone), and generalizes well to CFS and ABC datasets with different populations. These findings suggest that scEEG-PPG fusion is a promising approach for lightweight wearable based sleep monitoring, offering a pathway toward more accessible sleep health assessment. Source code of this project can be found at: https://github.com/DavyWJW/scEEG-PPGFusion
Executive Summary
This article explores the fusion of single-channel electroencephalography (scEEG) and photoplethysmography (PPG) for reliable sleep staging using lightweight wearables. The study investigates three fusion strategies and evaluates their performance on the Multi-Ethnic Study of Atherosclerosis (MESA) dataset, achieving a Cohen's Kappa of 0.798 and an accuracy of 86.9%. The findings suggest that scEEG-PPG fusion is a promising approach for sleep monitoring, offering a pathway toward more accessible sleep health assessment.
Key Points
- ▸ Combination of scEEG and PPG for sleep staging
- ▸ Evaluation of three fusion strategies: score-level fusion, cross-attention fusion, and Mamba-enhanced fusion
- ▸ Improvement in light sleep classification using scEEG-PPG fusion
Merits
Improved Accuracy
The scEEG-PPG fusion approach achieves higher accuracy and Cohen's Kappa values compared to using scEEG alone.
Generalizability
The approach generalizes well to different datasets, including the Cleveland Family Study (CFS) and the Apnea, Bariatric surgery, and CPAP (ABC) datasets.
Demerits
Limited Temporal Context
The study relies on short-window constraints (30 s - 30 min), which may not capture the full range of sleep patterns.
Dataset Limitations
The study uses a limited number of datasets, which may not be representative of the broader population.
Expert Commentary
The fusion of scEEG and PPG signals offers a promising approach for sleep staging, particularly in the context of lightweight wearables. The study's findings highlight the importance of considering the temporal context of sleep patterns and the need for more accurate and reliable algorithms for sleep staging. However, further research is needed to fully explore the potential of this approach and to address the limitations of the current study, including the limited temporal context and dataset limitations. Ultimately, the development of more accurate and reliable sleep monitoring systems has the potential to improve our understanding of sleep disorders and to inform the development of more effective treatments.
Recommendations
- ✓ Further research is needed to explore the potential of scEEG-PPG fusion for sleep staging in different populations and contexts.
- ✓ The development of more accurate and reliable algorithms for sleep staging should be a priority, particularly in the context of wearable technology and clinical applications.
