Automatic sleep staging and atonia detection in RBD using single-channel chin EMG

Area of Science:

• Neurology
• Sleep Medicine
• Biomedical Engineering

Background:

• Rapid eye movement (REM) sleep behavior disorder (RBD) diagnosis is crucial due to its association with neurodegenerative diseases.
• Current diagnostic methods rely on labor-intensive polysomnography (PSG) and manual scoring of REM sleep without atonia (RSWA).
• Existing automated methods often require manual REM scoring or have limited accuracy in RBD populations.

Purpose of the Study:

• To develop and validate a fully automated system for REM sleep scoring and RSWA quantification using only chin electromyography (EMG).
• To assess the feasibility and performance of a single-sensor approach for RBD diagnosis.

Main Methods:

• Utilized a factorized score-based diffusion model on single-channel chin EMG data from 485 PSG recordings (30 with RBD).
• Automated REM sleep epoch scoring and subsequent RSWA quantification using the REM atonia index.
• Compared automated results against human-scored PSG data.

Main Results:

• The automated system achieved high performance in REM scoring (sensitivity ≥0.80, specificity ≥0.97, PPV ≥0.81).
• Automated RSWA quantification showed strong agreement with manual scoring (bias ≤0.03, LoA ≤0.13, Pearson's r ≥0.88).
• The system demonstrated comparable performance to semi-automatic methods.

Conclusions:

• Fully automated RSWA quantification via single-channel chin EMG is feasible and accurate.
• This approach enhances the accessibility of RBD screening, especially in ambulatory settings.
• The single-sensor system facilitates long-term monitoring of RSWA for improved patient management.