The objective of this study was to investigate the accuracy of an automated sleep-staging algorithm that is based on a single bipolar EEG signal. Three raters each manually scored the polysomnographic records from 44 patients referred for sleep evaluation. The mean positive percent agreement across subjects and stages was 72.6%, approximately 80% for stages wake (78.3%), N2 (80.9%) and N3 (78.1%), decreased slightly to 73.2% for REM, and dropped to 31.9% for N1. This study demonstrated that automated scoring of sleep obtained from a single-channel of forehead EEG results in agreement to majority manual scoring similar to that obtained from studies of manual inter-rater agreement. The benefit in assessing auto-staging accuracy with consensus agreement across multiple raters is most apparent in patients with OSA, and limited disagreements in patients on medications and those with compromised signal quality.
Download Full Article