Anti-correlation time-frequency synergy and dual-granularity fusion for sleep stage classification
- 1School of Information and Control Engineering, China University of Mining and Technology
- 2School of Mechanical and Electrical Engineering, China University of Mining and Technology
- 3Department of Pediatric, Yangpu District Shidong Hospital
Abstract
Introduction: Automated sleep stage classification is crucial for diagnosing sleep disorders and evaluating sleep quality. Single-channel EEG sleep staging using central electrodes has garnered significant interest due to its ability to detect a higher number of sleep spindles and its cost-effectiveness. Nevertheless, several challenges remain, such as the neglect of interdependencies among features, reduced model sensitivity to the dispersion across features, potential information leakage during the learning of transition rules, and class imbalance.
Methods: To address these challenges, this study specifically designs an automatic sleep staging network based on single-channel EEG. The network incorporates a multi-branch feature extraction module, which captures negatively correlated frequency features that exhibit significant variations during sleep transitions and promotes synergistic complementarity. Furthermore, a dual-granularity residual attention mechanism is employed to perform adaptive fusion along both temporal and channel dimensions based on statistical measures such as the mean and standard deviation. The causal transition-aware network ensures temporal consistency while facilitating effective learning of transition rules. The focal loss function is utilized to counteract class imbalance.
Results: Experiments on SleepEDF-20, SleepEDF-78 and SHHS datasets show that our method achieves the best performance with overall accuracy of 87.0%, 83.4% and 84.5%, respectively. For N1 stage classification which is more challenging, it reaches an accuracy of 57.5%, 52.5% and 49.3% on three datasets, respectively.
Discussion: The experimental results demonstrate the effectiveness of the proposed approach and its certain advantages over existing methods.
Related
- Strong correlation between amino acid frequency and codon degeneracy in genetic codes across all domains of life
- Codon-aware multi-scale feature fusion for metagenomic sequence classification
- Critical Flicker Fusion Frequency As An Experience-Restricted Constraint On Visual Temporal Resolution: What Does And Does Not Change It
- FSF-RTDETR: a frequency–spatial feature fusion network for eggplant fruit disease and pest detection in complex agricultural scenes
- Beyond Discrimination: Evaluating Probability Calibration in Radiomics-Image Fusion for Renal Cell Carcinoma Classification
BibTeX
@article{pan2026anticorrelation,
title = {{Anti-correlation time-frequency synergy and dual-granularity fusion for sleep stage classification}},
author = {Jie Pan and Xingxing Zhao and Xiaoyu Zou and Pengjun Zhao and Leilei Wang and Chengcheng Li},
journal = {Frontiers in Neuroscience},
year = {2026},
month = oct,
doi = {10.3389/fnins.2026.1944816},
url = {https://www.frontiersin.org/articles/10.3389/fnins.2026.1944816/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Anti-correlation time-frequency synergy and dual-granularity fusion for sleep stage classification
AU - Jie Pan
AU - Xingxing Zhao
AU - Xiaoyu Zou
AU - Pengjun Zhao
AU - Leilei Wang
AU - Chengcheng Li
PY - 2026
DA - 2026/10/20
JO - Frontiers in Neuroscience
DO - 10.3389/fnins.2026.1944816
UR - https://www.frontiersin.org/articles/10.3389/fnins.2026.1944816/full
AB - Introduction: Automated sleep stage classification is crucial for diagnosing sleep disorders and evaluating sleep quality. Single-channel EEG sleep staging using central electrodes has garnered significant interest due to its ability to detect a higher number of sleep spindles and its cost-effectiveness. Nevertheless, several challenges remain, such as the neglect of interdependencies among features, reduced model sensitivity to the dispersion across features, potential information leakage during the learning of transition rules, and class imbalance. Methods: To address these challenges, this study specifically designs an automatic sleep staging network based on single-channel EEG. The network incorporates a multi-branch feature extraction module, which captures negatively correlated frequency features that exhibit significant variations during sleep transitions and promotes synergistic complementarity. Furthermore, a dual-granularity residual attention mechanism is employed to perform adaptive fusion along both temporal and channel dimensions based on statistical measures such as the mean and standard deviation. The causal transition-aware network ensures temporal consistency while facilitating effective learning of transition rules. The focal loss function is utilized to counteract class imbalance. Results: Experiments on SleepEDF-20, SleepEDF-78 and SHHS datasets show that our method achieves the best performance with overall accuracy of 87.0%, 83.4% and 84.5%, respectively. For N1 stage classification which is more challenging, it reaches an accuracy of 57.5%, 52.5% and 49.3% on three datasets, respectively. Discussion: The experimental results demonstrate the effectiveness of the proposed approach and its certain advantages over existing methods.
SN - 1662-453X
SN - 1662-4548
ER -