Evidence suggests Rapid-Eye-Movement (REM) Sleep Behaviour Disorder (RBD) is\nan early predictor of Parkinson's disease. This study proposes a\nfully-automated framework for RBD detection consisting of automated sleep\nstaging followed by RBD identification. Analysis was assessed using a limited\npolysomnography montage from 53 participants with RBD and 53 age-matched\nhealthy controls. Sleep stage classification was achieved using a Random Forest\n(RF) classifier and 156 features extracted from electroencephalogram (EEG),\nelectrooculogram (EOG) and electromyogram (EMG) channels. For RBD detection, a\nRF classifier was trained combining established techniques to quantify muscle\natonia with additional features that incorporate sleep architecture and the EMG\nfractal exponent. Automated multi-state sleep staging achieved a 0.62 Cohen's\nKappa score. RBD detection accuracy improved by 10% to 96% (compared to\nindividual established metrics) when using manually annotated sleep staging.\nAccuracy remained high (92%) when using automated sleep staging. This study\noutperforms established metrics and demonstrates that incorporating sleep\narchitecture and sleep stage transitions can benefit RBD detection. This study\nalso achieved automated sleep staging with a level of accuracy comparable to\nmanual annotation. This study validates a tractable, fully-automated, and\nsensitive pipeline for RBD identification that could be translated to wearable\ntake-home technology.\n