Diagnostic value of sleep stage dissociation as visualized on a 2-dimensional sleep state space in human narcolepsy

Anders Vinther Olsen; Jens Stephansen; Eileen Leary; Paul E Peppard; Hong Sheungshul; Poul Jørgen Jennum; Helge Sorensen; Emmanuel Mignot

doi:10.1016/j.jneumeth.2017.02.004

Diagnostic value of sleep stage dissociation as visualized on a 2-dimensional sleep state space in human narcolepsy

Anders Vinther Olsen, Jens Stephansen, Eileen Leary, Paul E Peppard, Hong Sheungshul, Poul Jørgen Jennum, Helge Sorensen, Emmanuel Mignot

Department of Clinical Medicine

8 Citations (Scopus)

Abstract

BACKGROUND: Type 1 narcolepsy (NT1) is characterized by symptoms believed to represent Rapid Eye Movement (REM) sleep stage dissociations, occurrences where features of wake and REM sleep are intermingled, resulting in a mixed state. We hypothesized that sleep stage dissociations can be objectively detected through the analysis of nocturnal Polysomnography (PSG) data, and that those affecting REM sleep can be used as a diagnostic feature for narcolepsy.

NEW METHOD: A Linear Discriminant Analysis (LDA) model using 38 features extracted from EOG, EMG and EEG was used in control subjects to select features differentiating wake, stage N1, N2, N3 and REM sleep. Sleep stage differentiation was next represented in a 2D projection. Features characteristic of sleep stage differences were estimated from the residual sleep stage probability in the 2D space. Using this model we evaluated PSG data from NT1 and non-narcoleptic subjects. An LDA classifier was used to determine the best separation plane.

COMPARISON WITH EXISTING METHODS: This method replicates the specificity/sensitivity from the training set to the validation set better than many other methods.

RESULTS: Eight prominent features could differentiate narcolepsy and controls in the validation dataset. Using a composite measure and a specificity cut off 95% in the training dataset, sensitivity was 43%. Specificity/sensitivity was 94%/38% in the validation set. Using hypersomnia subjects, specificity/sensitivity was 84%/15%. Analyzing treated narcoleptics the specificity/sensitivity was 94%/10%.

CONCLUSION: Sleep stage dissociation can be used for the diagnosis of narcolepsy. However the use of some medications and presence of undiagnosed hypersomnolence patients impacts the result.

Original language	English
Journal	Journal of Neuroscience Methods
Volume	282
Pages (from-to)	9-19
ISSN	0165-0270
DOIs	https://doi.org/10.1016/j.jneumeth.2017.02.004
Publication status	Published - 15 Apr 2017

Keywords

Adult
Cohort Studies
Discriminant Analysis
Electroencephalography/methods
Electromyography/methods
Electrooculography/methods
Humans
Linear Models
Machine Learning
Middle Aged
Narcolepsy/classification
Polysomnography/methods
Sensitivity and Specificity
Sleep Stages/drug effects

Access to Document

10.1016/j.jneumeth.2017.02.004

Cite this

@article{253a33eec9c047f8affc89df24a91cfa,

title = "Diagnostic value of sleep stage dissociation as visualized on a 2-dimensional sleep state space in human narcolepsy",

abstract = "BACKGROUND: Type 1 narcolepsy (NT1) is characterized by symptoms believed to represent Rapid Eye Movement (REM) sleep stage dissociations, occurrences where features of wake and REM sleep are intermingled, resulting in a mixed state. We hypothesized that sleep stage dissociations can be objectively detected through the analysis of nocturnal Polysomnography (PSG) data, and that those affecting REM sleep can be used as a diagnostic feature for narcolepsy.NEW METHOD: A Linear Discriminant Analysis (LDA) model using 38 features extracted from EOG, EMG and EEG was used in control subjects to select features differentiating wake, stage N1, N2, N3 and REM sleep. Sleep stage differentiation was next represented in a 2D projection. Features characteristic of sleep stage differences were estimated from the residual sleep stage probability in the 2D space. Using this model we evaluated PSG data from NT1 and non-narcoleptic subjects. An LDA classifier was used to determine the best separation plane.COMPARISON WITH EXISTING METHODS: This method replicates the specificity/sensitivity from the training set to the validation set better than many other methods.RESULTS: Eight prominent features could differentiate narcolepsy and controls in the validation dataset. Using a composite measure and a specificity cut off 95% in the training dataset, sensitivity was 43%. Specificity/sensitivity was 94%/38% in the validation set. Using hypersomnia subjects, specificity/sensitivity was 84%/15%. Analyzing treated narcoleptics the specificity/sensitivity was 94%/10%.CONCLUSION: Sleep stage dissociation can be used for the diagnosis of narcolepsy. However the use of some medications and presence of undiagnosed hypersomnolence patients impacts the result.",

keywords = "Adult, Cohort Studies, Discriminant Analysis, Electroencephalography/methods, Electromyography/methods, Electrooculography/methods, Humans, Linear Models, Machine Learning, Middle Aged, Narcolepsy/classification, Polysomnography/methods, Sensitivity and Specificity, Sleep Stages/drug effects",

author = "Olsen, {Anders Vinther} and Jens Stephansen and Eileen Leary and Peppard, {Paul E} and Hong Sheungshul and Jennum, {Poul J{\o}rgen} and Helge Sorensen and Emmanuel Mignot",

year = "2017",

month = apr,

day = "15",

doi = "10.1016/j.jneumeth.2017.02.004",

language = "English",

volume = "282",

pages = "9--19",

journal = "Journal of Neuroscience Methods",

issn = "0165-0270",

publisher = "Elsevier",

}

TY - JOUR

T1 - Diagnostic value of sleep stage dissociation as visualized on a 2-dimensional sleep state space in human narcolepsy

AU - Olsen, Anders Vinther

AU - Stephansen, Jens

AU - Leary, Eileen

AU - Peppard, Paul E

AU - Sheungshul, Hong

AU - Jennum, Poul Jørgen

AU - Sorensen, Helge

AU - Mignot, Emmanuel

PY - 2017/4/15

Y1 - 2017/4/15

N2 - BACKGROUND: Type 1 narcolepsy (NT1) is characterized by symptoms believed to represent Rapid Eye Movement (REM) sleep stage dissociations, occurrences where features of wake and REM sleep are intermingled, resulting in a mixed state. We hypothesized that sleep stage dissociations can be objectively detected through the analysis of nocturnal Polysomnography (PSG) data, and that those affecting REM sleep can be used as a diagnostic feature for narcolepsy.NEW METHOD: A Linear Discriminant Analysis (LDA) model using 38 features extracted from EOG, EMG and EEG was used in control subjects to select features differentiating wake, stage N1, N2, N3 and REM sleep. Sleep stage differentiation was next represented in a 2D projection. Features characteristic of sleep stage differences were estimated from the residual sleep stage probability in the 2D space. Using this model we evaluated PSG data from NT1 and non-narcoleptic subjects. An LDA classifier was used to determine the best separation plane.COMPARISON WITH EXISTING METHODS: This method replicates the specificity/sensitivity from the training set to the validation set better than many other methods.RESULTS: Eight prominent features could differentiate narcolepsy and controls in the validation dataset. Using a composite measure and a specificity cut off 95% in the training dataset, sensitivity was 43%. Specificity/sensitivity was 94%/38% in the validation set. Using hypersomnia subjects, specificity/sensitivity was 84%/15%. Analyzing treated narcoleptics the specificity/sensitivity was 94%/10%.CONCLUSION: Sleep stage dissociation can be used for the diagnosis of narcolepsy. However the use of some medications and presence of undiagnosed hypersomnolence patients impacts the result.

AB - BACKGROUND: Type 1 narcolepsy (NT1) is characterized by symptoms believed to represent Rapid Eye Movement (REM) sleep stage dissociations, occurrences where features of wake and REM sleep are intermingled, resulting in a mixed state. We hypothesized that sleep stage dissociations can be objectively detected through the analysis of nocturnal Polysomnography (PSG) data, and that those affecting REM sleep can be used as a diagnostic feature for narcolepsy.NEW METHOD: A Linear Discriminant Analysis (LDA) model using 38 features extracted from EOG, EMG and EEG was used in control subjects to select features differentiating wake, stage N1, N2, N3 and REM sleep. Sleep stage differentiation was next represented in a 2D projection. Features characteristic of sleep stage differences were estimated from the residual sleep stage probability in the 2D space. Using this model we evaluated PSG data from NT1 and non-narcoleptic subjects. An LDA classifier was used to determine the best separation plane.COMPARISON WITH EXISTING METHODS: This method replicates the specificity/sensitivity from the training set to the validation set better than many other methods.RESULTS: Eight prominent features could differentiate narcolepsy and controls in the validation dataset. Using a composite measure and a specificity cut off 95% in the training dataset, sensitivity was 43%. Specificity/sensitivity was 94%/38% in the validation set. Using hypersomnia subjects, specificity/sensitivity was 84%/15%. Analyzing treated narcoleptics the specificity/sensitivity was 94%/10%.CONCLUSION: Sleep stage dissociation can be used for the diagnosis of narcolepsy. However the use of some medications and presence of undiagnosed hypersomnolence patients impacts the result.

KW - Adult

KW - Cohort Studies

KW - Discriminant Analysis

KW - Electroencephalography/methods

KW - Electromyography/methods

KW - Electrooculography/methods

KW - Humans

KW - Linear Models

KW - Machine Learning

KW - Middle Aged

KW - Narcolepsy/classification

KW - Polysomnography/methods

KW - Sensitivity and Specificity

KW - Sleep Stages/drug effects

U2 - 10.1016/j.jneumeth.2017.02.004

DO - 10.1016/j.jneumeth.2017.02.004

M3 - Journal article

C2 - 28219726

SN - 0165-0270

VL - 282

SP - 9

EP - 19

JO - Journal of Neuroscience Methods

JF - Journal of Neuroscience Methods

ER -

Diagnostic value of sleep stage dissociation as visualized on a 2-dimensional sleep state space in human narcolepsy

Abstract

Keywords

Access to Document

Fingerprint

Cite this