A Context-Aware Temporal Modeling through Unified Multi-Scale Temporal Encoding and Hierarchical Sequence Learning for Single-Channel EEG Sleep Staging

2025-12-28

AI Summary

To address clinical deployment challenges in single-channel EEG-based automatic sleep staging—including poor N1-stage recognition, severe class imbalance, limited receptive fields, and low model interpretability—this paper proposes an interpretable, context-aware temporal modeling framework. Our method innovatively integrates unified multi-scale temporal encoding with a hierarchical sequence learning architecture, incorporates a sub-segment-level probability averaging mechanism to enhance contextual robustness, and employs class-weighted loss, time-domain data augmentation, and softmax fusion for improved generalization. Evaluated on the SleepEDF dataset, our approach achieves an overall accuracy of 89.72%, a macro-averaged F1-score of 85.46%, and an N1-stage F1-score of 61.7%—substantially outperforming state-of-the-art methods. The framework not only delivers high classification accuracy but also provides clinically meaningful interpretability through transparent temporal modeling and attention-guided feature aggregation.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

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 a context-aware and interpretable 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.The proposed model 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, classweighted 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 maintaining interpretability and suitability for real-world clinical applications.

Problem

• Improves N1 stage detection in single-channel EEG sleep staging
• Addresses class imbalance and limited receptive-field modeling challenges
• Enhances interpretability and robustness for clinical applications

Innovation

• Unified multi-scale temporal encoding for feature extraction
• Hierarchical sequence learning to capture long-range dependencies
• Class-weighted loss and data augmentation for imbalance handling