A Single-Channel EEG Approach for Sleep Stage-Independent Automatic Detection of REM Sleep Behavior Disorder

Rapid Eye Movement (REM) Sleep Behavior Disorder (RBD) is a parasomnia characterized by the loss of physiological muscle atonia during REM sleep, often manifesting through dream-enacting behavior. Idiopathic RBD is largely considered a prodromal stage of neurodegenerative diseases, with a conversion rate to overt α-synucleinopathies of up to 96% after 14 years. Currently, the diagnostic procedure relies on time-consuming and labor-intensive inspection of polysomnography (PSG). This study proposes a Machine Learning (ML), stage-agnostic framework for the automatic detection of RBD subjects through unstaged, single-channel EEG sleep data from 58 subjects (32 RBD). The best model achieved 86.21% accuracy, 90.6% sensitivity, and 87.9% F-1 score, demonstrating strong predictive power. This study is the first to explore whole-night EEG data for RBD detection, paving the way for scalable, lightweight clinical decision support systems for early neurodegenerative screening and risk assessment. This study presents a lightweight, clinical decision support tool to enhance RBD detection and support early interventions in neurodegenerative diseases.