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

Julie A E Christensen; Marielle Zoetmulder; Henriette Koch; Rune Frandsen; Lars Arvastson; Søren R Christensen; Poul Jennum; Helge B D Sorensen

doi:10.1016/j.jneumeth.2014.07.014

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

Julie A E Christensen, Marielle Zoetmulder, Henriette Koch, Rune Frandsen, Lars Arvastson, Søren R Christensen, Poul Jennum, Helge B D Sorensen

25 Citations (Scopus)

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.

Original language	English
Journal	Journal of Neuroscience Methods
Volume	235
Pages (from-to)	262–276
ISSN	0165-0270
DOIs	https://doi.org/10.1016/j.jneumeth.2014.07.014
Publication status	Published - 30 Sept 2014

Keywords

Aged
Algorithms
Artificial Intelligence
Electroencephalography
Electrooculography
Female
Humans
Male
Middle Aged
Models, Neurological
Nocturnal Myoclonus Syndrome
Parkinson Disease
Polysomnography
Regression Analysis
Sensitivity and Specificity
Signal Processing, Computer-Assisted
Sleep Stages

Access to Document

10.1016/j.jneumeth.2014.07.014

Cite this

@article{a48a271c95a64afc95d5c78a252ce38c,

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

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.",

keywords = "Aged, Algorithms, Artificial Intelligence, Electroencephalography, Electrooculography, Female, Humans, Male, Middle Aged, Models, Neurological, Nocturnal Myoclonus Syndrome, Parkinson Disease, Polysomnography, Regression Analysis, Sensitivity and Specificity, Signal Processing, Computer-Assisted, Sleep Stages",

author = "Christensen, {Julie A E} and Marielle Zoetmulder and Henriette Koch and Rune Frandsen and Lars Arvastson and Christensen, {S{\o}ren R} and Poul Jennum and Sorensen, {Helge B D}",

year = "2014",

month = sep,

day = "30",

doi = "10.1016/j.jneumeth.2014.07.014",

language = "English",

volume = "235",

pages = "262–276",

journal = "Journal of Neuroscience Methods",

issn = "0165-0270",

publisher = "Elsevier",

}

TY - JOUR

T1 - Data-driven modeling of sleep EEG and EOG reveals characteristics indicative of pre-Parkinson's and Parkinson's disease

AU - Christensen, Julie A E

AU - Zoetmulder, Marielle

AU - Koch, Henriette

AU - Frandsen, Rune

AU - Arvastson, Lars

AU - Christensen, Søren R

AU - Jennum, Poul

AU - Sorensen, Helge B D

PY - 2014/9/30

Y1 - 2014/9/30

N2 - 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.

AB - 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.

KW - Aged

KW - Algorithms

KW - Artificial Intelligence

KW - Electroencephalography

KW - Electrooculography

KW - Female

KW - Humans

KW - Male

KW - Middle Aged

KW - Models, Neurological

KW - Nocturnal Myoclonus Syndrome

KW - Parkinson Disease

KW - Polysomnography

KW - Regression Analysis

KW - Sensitivity and Specificity

KW - Signal Processing, Computer-Assisted

KW - Sleep Stages

U2 - 10.1016/j.jneumeth.2014.07.014

DO - 10.1016/j.jneumeth.2014.07.014

M3 - Journal article

C2 - 25088694

SN - 0165-0270

VL - 235

SP - 262

EP - 276

JO - Journal of Neuroscience Methods

JF - Journal of Neuroscience Methods

ER -

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

Abstract

Keywords

Access to Document

Fingerprint

Cite this