A context-aware temporal modeling through unified multi-scale temporal encoding and hierarchical sequence learning for single-channel EEG sleep staging
- 1Faculty of Computer Science and Engineering, Shahid Beheshti University, Tehran, Iran
Abstract
Automatic sleep staging is a critical task in healthcare due to the global prevalence of sleep disorders. This study focuses on single-channel electroencephalography (EEG), a practical and widely available signal for automatic sleep staging. Existing approaches face challenges such as class imbalance, limited receptive-field modeling, and insufficient interpretability. This work proposes TempoSleep, a context-aware framework for single-channel EEG sleep staging, with particular emphasis on improving detection of the N1 stage. Many prior models operate as black boxes with stacked layers, lacking clearly defined and interpretable feature extraction roles. TempoSleep combines compact multi-scale feature extraction with temporal modeling to capture both local and long-range dependencies. To address data imbalance, especially in the N1 stage, class-weighted loss functions and data augmentation are applied. EEG signals are segmented into sub-epoch chunks, and final predictions are obtained by averaging softmax probabilities across chunks, enhancing contextual representation and robustness. The proposed framework achieves an overall accuracy of 89.72% and a macro-average F1-score of 85.46%. Notably, it attains an F1-score of 61.7% for the challenging N1 stage, demonstrating a substantial improvement over previous methods on the SleepEDF datasets.These results indicate that the proposed approach effectively improves sleep staging performance while providing insights into its prediction behavior and supporting its potential application in automatic sleep staging.
Related
- UniHAR: A unified hierarchical transformer with multi-scale temporal modeling for wearable human activity recognition
- Event segmentation and encoding rely on information accumulated over different temporal scales
- Temporal and spatial alterations in GPi neuronal encoding might contribute to slow down movement in Parkinsonian monkeys
- Encoding and Decoding Temporal Signals with Spiking Bandpass Wavelets
- FLiTrak3D: Improved deep-learning-based 3D insect flightkinematics tracking using spatial and temporal encoding
BibTeX
@article{vakili2026contextaware,
title = {{A context-aware temporal modeling through unified multi-scale temporal encoding and hierarchical sequence learning for single-channel EEG sleep staging}},
author = {Amir Ali Vakili and Salar Jahanshiri and Armin Salimi-Badr},
journal = {PLOS One},
year = {2026},
month = sep,
doi = {10.1371/journal.pone.0358241},
url = {https://doi.org/10.1371/journal.pone.0358241},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A context-aware temporal modeling through unified multi-scale temporal encoding and hierarchical sequence learning for single-channel EEG sleep staging
AU - Amir Ali Vakili
AU - Salar Jahanshiri
AU - Armin Salimi-Badr
PY - 2026
DA - 2026/09/30
JO - PLOS One
DO - 10.1371/journal.pone.0358241
UR - https://doi.org/10.1371/journal.pone.0358241
AB - Automatic sleep staging is a critical task in healthcare due to the global prevalence of sleep disorders. This study focuses on single-channel electroencephalography (EEG), a practical and widely available signal for automatic sleep staging. Existing approaches face challenges such as class imbalance, limited receptive-field modeling, and insufficient interpretability. This work proposes TempoSleep, a context-aware framework for single-channel EEG sleep staging, with particular emphasis on improving detection of the N1 stage. Many prior models operate as black boxes with stacked layers, lacking clearly defined and interpretable feature extraction roles. TempoSleep combines compact multi-scale feature extraction with temporal modeling to capture both local and long-range dependencies. To address data imbalance, especially in the N1 stage, class-weighted loss functions and data augmentation are applied. EEG signals are segmented into sub-epoch chunks, and final predictions are obtained by averaging softmax probabilities across chunks, enhancing contextual representation and robustness. The proposed framework achieves an overall accuracy of 89.72% and a macro-average F1-score of 85.46%. Notably, it attains an F1-score of 61.7% for the challenging N1 stage, demonstrating a substantial improvement over previous methods on the SleepEDF datasets.These results indicate that the proposed approach effectively improves sleep staging performance while providing insights into its prediction behavior and supporting its potential application in automatic sleep staging.
SN - 1932-6203
ER -