Julie A E Christensen1, Marielle Zoetmulder2, Henriette Koch3, Rune Frandsen4, Lars Arvastson5, Søren R Christensen5, Poul Jennum6, Helge B D Sorensen3. 1. Department of Electrical Engineering, Technical University of Denmark, Orsteds Plads, Building 349, DK-2800 Kongens Lyngby, Denmark; Danish Center for Sleep Medicine, University of Copenhagen, Department of Clinical Neurophysiology, Glostrup Hospital, Entrance 5, Nordre Ringvej 57, DK-2600 Glostrup, Denmark; H. Lundbeck A/S, Ottiliavej 9, DK-2500 Valby, Denmark. Electronic address: julie.a.e.christensen@gmail.com. 2. Danish Center for Sleep Medicine, University of Copenhagen, Department of Clinical Neurophysiology, Glostrup Hospital, Entrance 5, Nordre Ringvej 57, DK-2600 Glostrup, Denmark; Department of Neurology, Bispebjerg Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen, Denmark. 3. Department of Electrical Engineering, Technical University of Denmark, Orsteds Plads, Building 349, DK-2800 Kongens Lyngby, Denmark. 4. Danish Center for Sleep Medicine, University of Copenhagen, Department of Clinical Neurophysiology, Glostrup Hospital, Entrance 5, Nordre Ringvej 57, DK-2600 Glostrup, Denmark. 5. H. Lundbeck A/S, Ottiliavej 9, DK-2500 Valby, Denmark. 6. Danish Center for Sleep Medicine, University of Copenhagen, Department of Clinical Neurophysiology, Glostrup Hospital, Entrance 5, Nordre Ringvej 57, DK-2600 Glostrup, Denmark; Center for Healthy Ageing, University of Copenhagen, Blegdamsvej 3B, DK-2200 Copenhagen N, Denmark.
Abstract
BACKGROUND: Manual scoring of sleep relies on identifying certain characteristics in polysomnograph (PSG) signals. However, these characteristics are disrupted in patients with neurodegenerative diseases. NEW METHOD: 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. RESULTS: 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. COMPARISON WITH EXISTING METHOD: The topics showed visual accordance with the manually scored sleep stages, and the features revealed sleep characteristics containing information indicative of neurodegeneration. CONCLUSIONS: 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.
BACKGROUND: Manual scoring of sleep relies on identifying certain characteristics in polysomnograph (PSG) signals. However, these characteristics are disrupted in patients with neurodegenerative diseases. NEW METHOD: 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. RESULTS: 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/PDpatients. COMPARISON WITH EXISTING METHOD: The topics showed visual accordance with the manually scored sleep stages, and the features revealed sleep characteristics containing information indicative of neurodegeneration. CONCLUSIONS: 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 neurodegenerativepatients.
Authors: Michele A Bertocci; Genna Bebko; Amanda Dwojak; Satish Iyengar; Cecile D Ladouceur; Jay C Fournier; Amelia Versace; Susan B Perlman; Jorge R C Almeida; Michael J Travis; Mary Kay Gill; Lisa Bonar; Claudiu Schirda; Vaibhav A Diwadkar; Jeffrey L Sunshine; Scott K Holland; Robert A Kowatch; Boris Birmaher; David Axelson; Sarah M Horwitz; Thomas Frazier; L Eugene Arnold; Mary A Fristad; Eric A Youngstrom; Robert L Findling; Mary L Phillips Journal: Biol Psychiatry Cogn Neurosci Neuroimaging Date: 2017-05
Authors: M A Bertocci; G Bebko; A Versace; S Iyengar; L Bonar; E E Forbes; J R C Almeida; S B Perlman; C Schirda; M J Travis; M K Gill; V A Diwadkar; J L Sunshine; S K Holland; R A Kowatch; B Birmaher; D A Axelson; T W Frazier; L E Arnold; M A Fristad; E A Youngstrom; S M Horwitz; R L Findling; M L Phillips Journal: Psychol Med Date: 2016-12-21 Impact factor: 7.723
Authors: Julie A E Christensen; Miki Nikolic; Simon C Warby; Henriette Koch; Marielle Zoetmulder; Rune Frandsen; Keivan K Moghadam; Helge B D Sorensen; Emmanuel Mignot; Poul J Jennum Journal: Front Hum Neurosci Date: 2015-05-01 Impact factor: 3.169
Authors: M A Bertocci; G Bebko; A Versace; J C Fournier; S Iyengar; T Olino; L Bonar; J R C Almeida; S B Perlman; C Schirda; M J Travis; M K Gill; V A Diwadkar; E E Forbes; J L Sunshine; S K Holland; R A Kowatch; B Birmaher; D Axelson; S M Horwitz; T W Frazier; L E Arnold; M A Fristad; E A Youngstrom; R L Findling; M L Phillips Journal: Mol Psychiatry Date: 2016-02-23 Impact factor: 15.992
Authors: Alexander J Baumgartner; Clete A Kushida; Michael O Summers; Drew S Kern; Aviva Abosch; John A Thompson Journal: Front Neurol Date: 2021-10-28 Impact factor: 4.003