SleePyCo: Automatic Sleep Scoring with Feature Pyramid and Contrastive Learning

20 Sep 2022  ·  Seongju Lee, Yeonguk Yu, Seunghyeok Back, Hogeon Seo, Kyoobin Lee ·

Automatic sleep scoring is essential for the diagnosis and treatment of sleep disorders and enables longitudinal sleep tracking in home environments. Conventionally, learning-based automatic sleep scoring on single-channel electroencephalogram (EEG) is actively studied because obtaining multi-channel signals during sleep is difficult. However, learning representation from raw EEG signals is challenging owing to the following issues: 1) sleep-related EEG patterns occur on different temporal and frequency scales and 2) sleep stages share similar EEG patterns. To address these issues, we propose a deep learning framework named SleePyCo that incorporates 1) a feature pyramid and 2) supervised contrastive learning for automatic sleep scoring. For the feature pyramid, we propose a backbone network named SleePyCo-backbone to consider multiple feature sequences on different temporal and frequency scales. Supervised contrastive learning allows the network to extract class discriminative features by minimizing the distance between intra-class features and simultaneously maximizing that between inter-class features. Comparative analyses on four public datasets demonstrate that SleePyCo consistently outperforms existing frameworks based on single-channel EEG. Extensive ablation experiments show that SleePyCo exhibits enhanced overall performance, with significant improvements in discrimination between the N1 and rapid eye movement (REM) stages.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Sleep Stage Detection MASS (single-channel) SleePyCo (C4-A1 only) Accuracy 86.8% # 1
Cohen's Kappa 0.811 # 1
Macro-F1 0.825 # 1
Sleep Stage Detection Montreal Archive of Sleep Studies SleePyCo (C4-A1 only) Accuracy 86.8% # 1
Cohen's kappa 0.811 # 1
Macro-F1 0.825 # 1
Sleep Stage Detection PhysioNet Challenge 2018 SleePyCo (C3-A2 only) Accuracy 80.9% # 2
Cohen's Kappa 0.737 # 2
Macro-F1 0.789 # 2
Sleep Stage Detection PhysioNet Challenge 2018 (single-channel) SleePyCo (C3-A2 only) Accuracy 80.9% # 1
Cohen's Kappa 0.737 # 1
Macro-F1 0.789 # 1
Sleep Stage Detection SHHS SleePyCo (C4-A1 only) Accuracy 87.9% # 4
Cohen's Kappa 0.830 # 4
Macro-F1 0.807 # 4
Sleep Stage Detection SHHS (single-channel) SleePyCo (C4-A1 only) Accuracy 87.9% # 1
Cohen's Kappa 0.830 # 1
Macro-F1 0.807 # 1
Sleep Stage Detection Sleep-EDF SleePyCo (Fpz-Cz only) Accuracy 86.8% # 1
Macro-F1 0.812 # 1
Cohen's kappa 0.820 # 1
Sleep Stage Detection Sleep-EDF (single-channel) SleePyCo (Fpz-Cz only) Accuracy 86.8% # 1
Sleep Stage Detection Sleep-EDFx SleePyCo (Fpz-Cz only) Accuracy 84.6% # 1
Cohen's Kappa 0.787 # 1
Macro-F1 0.790 # 1
Sleep Stage Detection Sleep-EDFx (single-channel) SleePyCo (Fpz-Cz only) Accuracy 84.6% # 1
Cohen's Kappa 0.787 # 1
Macro-F1 0.790 # 1

Methods