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

cs.LGFeb 27, 2026

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

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

Organizations: School of Computer Science and Technology, Dalian University of Technology, Dalian, China · Faculty of Information Technology, University of Jyväskylä, Jyväskylä, Finland · Key Laboratory of Social Computing and Cognitive Intelligence, Ministry of Education, Dalian, China · School of Biomedical Engineering, Faculty of Medicine, Dalian University of Technology, Dalian, China · Department of Computing Sciences, Tampere University, Tampere, Finland · Faculty of Medicine, University of Helsinki, Helsinki, Finland

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.

Figures & tables

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.

Ablated Component	Configuration	ACC (%)	Params	FLOPs
Proposed (ULW-SleepNet)	3 blocks, ks = 3, ps = 2, fs = (8, 16, 32), SeparableConv1D, EEG/EOG/EMG	86.88	13,337	7.89 M
Input modality	only EEG	85.58	13,337	7.89 M
DSSC Block Number	2 blocks, fs = (16, 32)	85.37	12,841	17.39 M
DSSC Block Number	4 blocks, fs = (8, 16, 32, 64)	86.32	34,713	10.39 M
Kernel Size & Pooling Stride	ks = 7, ps = 4	86.57	13,661	9.19 M
Filter Number	fs = (4, 8, 16)	84.64	5,873	2.54 M

Table 1: Ablation study on the Sleep-EDF-20 dataset.

Dataset	Model	Overall Performance			Per-stage F1 Score (%)					Resource
Dataset	Model	ACC (%)	MF1 (%)	$\kappa$	Wake	N1	N2	N3	REM	Params	FLOPs
Sleep-EDF-20	DeepSleepNet [ 14 ]	82.0	76.9	0.76	84.7	46.6	85.9	84.8	82.4	24.7M	–
	DeepSleepNet-lite [ 4 ]	84.0	78.0	0.78	87.1	44.4	87.9	88.2	82.4	600K	–
	TinySleepNet [ 15 ]	85.4	80.5	0.80	90.1	51.4	88.5	88.3	84.3	1.3M	–
	AttnSleepNet [ 3 ]	84.4	78.1	0.79	89.7	42.6	88.8	90.2	79.0	520K	60.9M
	LWSleepNet [ 19 ]	86.6	79.2	0.81	92.4	41.3	90.2	88.4	84.0	180K	55.3M

Table 2: Comparison with SOTA models on two datasets.

Explore similar work

AnySleep: a channel-agnostic deep learning system for high-resolution sleep staging in multi-center cohorts
Dec 16, 2025Niklas Grieger, Jannik Raskob, Siamak Mehrkanoon +1Electroencephalography Sleep Stage Classification
Department of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences, 52428 Jülich, Germany · Department of Information and Computing Sciences, Utrecht University, Utrecht, The Netherlands · Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences, 52428 Jülich, Germany
NanoSleep: A Parameter-Efficient Hybrid Temporal Convolutional Network for Single-Channel Sleep Stage Classification
Aug 19, 2026S M Asif Hossain, Shruti KshirsagarSleep Stage Classification Lightweight CNNs
School of Computing, Wichita State University, Kansas, USA
Lightweight ML-Based Automatic Sleep Staging Framework with Constrained CNN and Mamba for Small-Sample EEG Datasets
Jul 6, 2026Zihao Wei, Yulin Gong, Yudan LvEfficient Neural Network Inference Electroencephalography
School of Electronic Information Engineering, Changchun University of Science and Technology, Changchun 130022, Jilin Province, China · Jilin University, Changchun 130012, Jilin Province, China

Dec 16, 2025J/K move · Enter open · S save

AnySleep: a channel-agnostic deep learning system for high-resolution sleep staging in multi-center cohorts

Niklas Grieger, Jannik Raskob, Siamak Mehrkanoon, Stephan Bialonski

Department of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences, 52428 Jülich, Germany · Department of Information and Computing Sciences, Utrecht University, Utrecht, The Netherlands · Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences, 52428 Jülich, Germany

Electroencephalography Sleep Stage Classification EEG Decoding Domain Generalization

Open paper AskPDF arXiv

Sleep is essential for health, yet studying its dynamics requires manual sleep staging, a labor-intensive step in research and clinical care. Across centers, polysomnography (PSG) recordings are traditionally scored in 30-s epochs for pragmatic, not physiological, reasons and vary in electrode count, montage, and subject characteristics. These constraints challenge harmonized multi-center studies and the discovery of robust biomarkers on shorter timescales. We present AnySleep, a deep neural network that scores sleep from any electroencephalography (EEG) or electrooculography (EOG) data at adjustable temporal resolutions. We trained and validated the model on over 20,000 overnight recordings (> 200,000 hours of EEG and EOG) from 28 datasets across multiple clinics to promote robust generalization across sites. The model attains state-of-the-art performance and surpasses or equals established baselines at 30-s epochs. Performance improves with more channels, yet remains strong when EOG is absent or only EOG or single EEG derivations (frontal, central, or occipital) are available. On sub-30-s timescales, the model captures short wake intrusions consistent with arousals and improves prediction of pathophysiological conditions (obstructive sleep apnea, narcolepsy type 1, insomnia) over 30-s scoring. We make the model publicly available to facilitate large-scale studies with heterogeneous electrode setups and accelerate biomarker discovery in sleep.

Abstract

Explore similar work

AnySleep: a channel-agnostic deep learning system for high-resolution sleep staging in multi-center cohorts

NanoSleep: A Parameter-Efficient Hybrid Temporal Convolutional Network for Single-Channel Sleep Stage Classification

Lightweight ML-Based Automatic Sleep Staging Framework with Constrained CNN and Mamba for Small-Sample EEG Datasets