Automatic sleep staging and atonia detection in RBD using single-channel chin EMG
Hans Van Gorp, Jaap F Van Der Aar, Ruud JG Van Sloun, Angelique Pijpers, Pedro Fonseca, Sebastiaan Overeem, Merel M Van GilstStudy Objectives
The diagnosis of rapid eye movement (REM) sleep behavior disorder (RBD) is of clinical interest, as there is a high rate of future conversion to Parkinson's disease and related neurodegenerative disorders. The clinical standard for diagnosis of RBD relies on human scoring of REM sleep without atonia (RSWA) via polysomnography, which is labor-intensive, costly, and unsuitable for long-term monitoring. Existing (semi-)automatic detection methods often depend on prior manual annotation of REM or suffer from low REM classification performance in this population. We here propose a fully automated system using only a single-channel chin electromyography (EMG) for both REM scoring and RSWA quantification.
Methods
We analyzed 485 polysomnographic recordings, including 30 recordings of subjects with RBD. Using the recently introduced factorized score-based diffusion model, we automatically scored REM epochs based on single-channel chin EMG data, and subsequently quantified RSWA using the REM atonia index from literature.
Results
The proposed system demonstrated strong performance in REM scoring in both subjects with and without RBD (REM sensitivity ≥0.80, specificity ≥0.97, positive predictive value ≥0.81). The REM atonia index derived from automatic scoring showed strong agreement with manually scored data, with minimal bias (≤0.03), narrow limits of agreement (≤0.13), and strong linear correlation (Peason's r ≥ 0.88).
Conclusions
Fully automated RSWA quantification using single-channel chin EMG is feasible and performs comparable to semi-automatic methods that rely on human sleep stage scoring based on polysomnography. This single-sensor approach can improve RBD screening accessibility, particularly in ambulatory settings, and facilitate long term monitoring of RSWA.