Objective: To facilitate quantification of REM sleep without atonia (RSWA) and thereby identify patients with rapid eye movement (REM) sleep behaviour disorder (RBD).

Background: RBD is defined by dream enactment and RSWA. We introduce RBDtector, an open-source software that allows to automatically classify RSWA according to established criteria.

Method: Two raters manually scored RSWA in ten participants with and ten participants without RBD as tonic, phasic, and any muscle activity according to the SINBAR criteria from polysomnograms encompassing surface EMG (electromyography) of the mentalis, the flexor digitorum superficialis (FDS), and the anterior tibialis (AT) muscles. The same data was then analysed using RBDtector. To examine its performance, datasets of 174 additional participants (102 with RBD and 72 without RBD) were investigated with RBDtector.

Results: RBDtector achieved a high level of agreement with human ratings, particularly with an expert rater. The highest congruency was detected for phasic and any activity of the FDS. RBDtector was able to identify RBD-positive participants with 100% specificity and 95.7% sensitivity after discarding artifacts when mentalis phasic and FDS any activity were combined and a cut-off of 20.6 % RSWA was used. Omitting manual artifact removal led to a similar performance. RBD-positive subjects had muscle bouts of higher amplitude and longer duration.

Conclusion: The open-source software RBDtector enables to automatically score RSWA with high inter-rater reliability with an expert human rater and can thereby help to identify participants with RBD.

« Back to 2022 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/automated-scoring-of-rem-sleep-without-atonia-using-the-open-source-software-rbdtector/