Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ULW-SleepNet: An Ultra-Lightweight Network for Multimodal Sleep Stage Scoring

Zhaowen Wang, Dongdong Zhou, Qi Xu, Fengyu Cong, Mohammad Al-Sa'd, Jenni Raitoharju

TL;DR

ULW-SleepNet is proposed, an ultra-lightweight multimodal sleep stage scoring framework that efficiently integrates information from multiple physiological signals and reduces parameters by up to 98.6% with only marginal performance loss, demonstrating its strong potential for real-time sleep monitoring on wearable and IoT devices.

Abstract

Automatic sleep stage scoring is crucial for the diagnosis and treatment of sleep disorders. Although deep learning models have advanced the field, many existing models are computationally demanding and designed for single-channel electroencephalography (EEG), limiting their practicality for multimodal polysomnography (PSG) data. To overcome this, we propose ULW-SleepNet, an ultra-lightweight multimodal sleep stage scoring framework that efficiently integrates information from multiple physiological signals. ULW-SleepNet incorporates a novel Dual-Stream Separable Convolution (DSSC) Block, depthwise separable convolutions, channel-wise parameter sharing, and global average pooling to reduce computational overhead while maintaining competitive accuracy. Evaluated on the Sleep-EDF-20 and Sleep-EDF-78 datasets, ULW-SleepNet achieves accuracies of 86.9% and 81.4%, respectively, with only 13.3K parameters and 7.89M FLOPs. Compared to state-of-the-art methods, our model reduces parameters by up to 98.6% with only marginal performance loss, demonstrating its strong potential for real-time sleep monitoring on wearable and IoT devices. The source code for this study is publicly available at https://github.com/wzw999/ULW-SLEEPNET.

ULW-SleepNet: An Ultra-Lightweight Network for Multimodal Sleep Stage Scoring

TL;DR

ULW-SleepNet is proposed, an ultra-lightweight multimodal sleep stage scoring framework that efficiently integrates information from multiple physiological signals and reduces parameters by up to 98.6% with only marginal performance loss, demonstrating its strong potential for real-time sleep monitoring on wearable and IoT devices.

Abstract

Automatic sleep stage scoring is crucial for the diagnosis and treatment of sleep disorders. Although deep learning models have advanced the field, many existing models are computationally demanding and designed for single-channel electroencephalography (EEG), limiting their practicality for multimodal polysomnography (PSG) data. To overcome this, we propose ULW-SleepNet, an ultra-lightweight multimodal sleep stage scoring framework that efficiently integrates information from multiple physiological signals. ULW-SleepNet incorporates a novel Dual-Stream Separable Convolution (DSSC) Block, depthwise separable convolutions, channel-wise parameter sharing, and global average pooling to reduce computational overhead while maintaining competitive accuracy. Evaluated on the Sleep-EDF-20 and Sleep-EDF-78 datasets, ULW-SleepNet achieves accuracies of 86.9% and 81.4%, respectively, with only 13.3K parameters and 7.89M FLOPs. Compared to state-of-the-art methods, our model reduces parameters by up to 98.6% with only marginal performance loss, demonstrating its strong potential for real-time sleep monitoring on wearable and IoT devices. The source code for this study is publicly available at https://github.com/wzw999/ULW-SLEEPNET.
Paper Structure (16 sections, 1 figure, 2 tables)

This paper contains 16 sections, 1 figure, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. ULW-SleepNet Architecture Overview
  4. Key Lightweight Design Strategies
  5. Depthwise Separable Convolution Integration
  6. Channel-Wise Processing with Parameter Sharing
  7. DSSC Block Architecture
  8. Global Average Pooling for Dimension Reduction
  9. Model Configuration
  10. Experiments and Results
  11. Datasets
  12. Experimental setting
  13. Ablation study
  14. Comparison with SOTA methods
  15. Conclusion and future work
  16. ...and 1 more sections

Figures (1)

  • Figure 1: An overview of the ULW-SleepNet architecture. Multimodal physiological input signals (i.e., EEG, EOG, EMG) are processed through a shared channel-wise feature extraction pipeline based on the Dual-Stream Separable Convolutional Block.