Data-driven modeling of sleep EEG and EOG reveals characteristics indicative of pre-Parkinson's and Parkinson's disease

• Published in: Journal of neuroscience methods 2014-09, Vol.235, p.262-276
• Main Authors: Christensen, Julie A.E., Zoetmulder, Marielle, Koch, Henriette, Frandsen, Rune, Arvastson, Lars, Christensen, Søren R., Jennum, Poul, Sorensen, Helge B.D.
• Format: Article
• Language: English
• Subjects: Aged; Algorithms; Artificial Intelligence; Automatic classification; Electroencephalography; Electroencephalography - methods; Electrooculography; Electrooculography - methods; Female; Humans; Male; Middle Aged; Models, Neurological; Nocturnal Myoclonus Syndrome - diagnosis; Nocturnal Myoclonus Syndrome - physiopathology; Parkinson Disease - diagnosis; Parkinson Disease - physiopathology; Parkinson's disease; Polysomnography; Polysomnography - methods; Regression Analysis; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Sleep Stages - physiology; Topic modeling
• ISSN: 0165-0270; 1872-678X
• DOI: 10.1016/j.jneumeth.2014.07.014

•A data-driven topic modeling approach characterizing sleep EEG and EOG is proposed.•The approach showed potential for evaluating patients with neurodegeneration.•The number of topics linked with REM and N3 could be an early PD biomarker.•The ability to maintain NREM and REM sleep could be an early PD biomarker.•Patients were classified with 91.4% sensitivity and 68.8% specificity. Manual scoring of sleep relies on identifying certain characteristics in polysomnograph (PSG) signals. However, these characteristics are disrupted in patients with neurodegenerative diseases. This study evaluates sleep using a topic modeling and unsupervised learning approach to identify sleep topics directly from electroencephalography (EEG) and electrooculography (EOG). PSG data from control subjects were used to develop an EOG and an EEG topic model. The models were applied to PSG data from 23 control subjects, 25 patients with periodic leg movements (PLMs), 31 patients with idiopathic REM sleep behavior disorder (iRBD) and 36 patients with Parkinson's disease (PD). The data were divided into training and validation datasets and features reflecting EEG and EOG characteristics based on topics were computed. The most discriminative feature subset for separating iRBD/PD and PLM/controls was estimated using a Lasso-regularized regression model. The features with highest discriminability were the number and stability of EEG topics linked to REM and N3, respectively. Validation of the model indicated a sensitivity of 91.4% and a specificity of 68.8% when classifying iRBD/PD patients. The topics showed visual accordance with the manually scored sleep stages, and the features revealed sleep characteristics containing information indicative of neurodegeneration. This study suggests that the amount of N3 and the ability to maintain NREM and REM sleep have potential as early PD biomarkers. Data-driven analysis of sleep may contribute to the evaluation of neurodegenerative patients.