Deep Learning Approaches for Sleep Apnea Classification from Polysomnographic EEG Signals

2026-07-16

AI Summary

This study addresses the high cost and reliance on manual polysomnography analysis in diagnosing pediatric obstructive sleep apnea (OSA) by exploring an automated screening approach based solely on electroencephalography (EEG) signals. Within a unified pediatric cohort, the authors systematically evaluate the performance of Vision Transformer and Graph Attention Network architectures across multiple EEG representations—including raw time-series, spectrograms, coherence maps, and features derived from topological data analysis (TDA)—introducing TDA for the first time to enhance discriminative capacity. Experiments involving 575 children achieved a maximum AUC of 0.750, demonstrating the feasibility of single-modality EEG-based automated OSA detection. The study further reveals significant heterogeneity in model performance across subgroups defined by age, sex, apnea–hypopnea index severity, and sleep stage.

Abstract

Sleep apnea diagnosis via polysomnography remains resource intensive and relies on time consuming manual data analysis and scoring. Recent work has demonstrated that central nervous system effects of sleep apnea events can be detected through electroencephalogram (EEG) signals. However, most work uses a single feature type on various datasets combined with different classification algorithms. In this work, we present a comprehensive comparison of deep learning architectures and feature representations for automated sleep apnea detection from multichannel EEG on a single dataset of pediatric subjects. We evaluate Vision Transformers and Graph Attention Networks across distinct signal representations: raw temporal signals, short-time Fourier transform spectrograms, coherence based graphs, and two topological data analysis (TDA) derived features. Using age and sex matching of our train and test sets, we train on 2410 pediatric subjects and test on 575 pediatric subjects. We achieve a best test AUC of 0.750 using a vision transformer based model trained on TDA features. Stratified analysis across patient demographics (age, sex, AHI severity) and sleep stages (N1, N2, N3, REM) reveals significant performance variation. Our results demonstrate the feasibility of EEG based automated OSA screening while highlighting essential challenges for clinical deployment.

Problem

• sleep apnea
• EEG signals
• automated detection
• polysomnography
• pediatric

Innovation

• Vision Transformer
• Topological Data Analysis (TDA)
• Graph Attention Network
• EEG-based sleep apnea detection
• Multichannel EEG representation