Seizure detection in adult ICU patients based on changes in EEG synchronization likelihood

A J C Slooter; E M Vriens; F S S Leijten; J J Spijkstra; A R J Girbes; A C van Huffelen; C J Stam

doi:https://doi.org/10.1385/NCC:5:3:186

Seizure detection in adult ICU patients based on changes in EEG synchronization likelihood

A J C Slooter, E M Vriens, F S S Leijten, J J Spijkstra, A R J Girbes, A C van Huffelen, C J Stam

Research output: Contribution to journal › Article › Academic › peer-review

28 Citations (Scopus)

Abstract

INTRODUCTION: Seizures are common in Intensive Care Unit (ICU) patients, and may increase neuronal injury.

PURPOSE: To explore the possible value of synchronization likelihood (SL) for the automatic detection of seizures in adult ICU patients.

METHODS: We included EEGs from ICU patients with a variety of diagnoses. The gold standard for further analyses was the consensus judgment of three clinical neurophysiologists who classified 150 scalp EEG epochs as "definitely epileptiform," "definitely non epileptiform," or "uncertain." SL estimates the statistical interdependencies between two time series, such as two EEG channels. We computed the average synchronization by calculating the SL between one channel and every other channel, and taking the mean of these values.

RESULTS: The mean SL in the 38 "definitely epileptiform" epochs ranged from 0.095 to 0.386 (mean 0.189; SD 0.066). In the 34 "definitely nonepileptiform" epochs the mean SL ranged from 0.087 to 0.158 (mean 0.115; SD 0.016; p < 0.0005). The area under the ROC curve was 0.812 (95% Confidence Interval 0.725 to 0.898).

CONCLUSION: The mean SL may distinguish between seizure and nonseizure epochs, and may prove helpful to monitor epileptic activity in ICU patients.

Original language	English
Pages (from-to)	186-92
Number of pages	7
Journal	Neurocritical Care
Volume	5
Issue number	3
DOIs	https://doi.org/10.1385/NCC:5:3:186
Publication status	Published - 2006

Keywords

Cerebral Cortex/physiopathology
Cortical Synchronization
Diagnosis, Differential
Electroencephalography/statistics & numerical data
Epilepsy/diagnosis
Evoked Potentials/physiology
Humans
Intensive Care Units
Likelihood Functions
Nonlinear Dynamics
Observer Variation
ROC Curve
Signal Processing, Computer-Assisted
Status Epilepticus/diagnosis

Access to Document

https://doi.org/10.1385/NCC:5:3:186

Cite this

@article{97c5b9b25fcd4ff8973346d18cffc40d,

title = "Seizure detection in adult ICU patients based on changes in EEG synchronization likelihood",

abstract = "INTRODUCTION: Seizures are common in Intensive Care Unit (ICU) patients, and may increase neuronal injury.PURPOSE: To explore the possible value of synchronization likelihood (SL) for the automatic detection of seizures in adult ICU patients.METHODS: We included EEGs from ICU patients with a variety of diagnoses. The gold standard for further analyses was the consensus judgment of three clinical neurophysiologists who classified 150 scalp EEG epochs as {"}definitely epileptiform,{"} {"}definitely non epileptiform,{"} or {"}uncertain.{"} SL estimates the statistical interdependencies between two time series, such as two EEG channels. We computed the average synchronization by calculating the SL between one channel and every other channel, and taking the mean of these values.RESULTS: The mean SL in the 38 {"}definitely epileptiform{"} epochs ranged from 0.095 to 0.386 (mean 0.189; SD 0.066). In the 34 {"}definitely nonepileptiform{"} epochs the mean SL ranged from 0.087 to 0.158 (mean 0.115; SD 0.016; p < 0.0005). The area under the ROC curve was 0.812 (95% Confidence Interval 0.725 to 0.898).CONCLUSION: The mean SL may distinguish between seizure and nonseizure epochs, and may prove helpful to monitor epileptic activity in ICU patients.",

keywords = "Cerebral Cortex/physiopathology, Cortical Synchronization, Diagnosis, Differential, Electroencephalography/statistics & numerical data, Epilepsy/diagnosis, Evoked Potentials/physiology, Humans, Intensive Care Units, Likelihood Functions, Nonlinear Dynamics, Observer Variation, ROC Curve, Signal Processing, Computer-Assisted, Status Epilepticus/diagnosis",

author = "Slooter, {A J C} and Vriens, {E M} and Leijten, {F S S} and Spijkstra, {J J} and Girbes, {A R J} and {van Huffelen}, {A C} and Stam, {C J}",

year = "2006",

doi = "https://doi.org/10.1385/NCC:5:3:186",

language = "English",

volume = "5",

pages = "186--92",

journal = "Neurocritical Care",

issn = "1541-6933",

publisher = "Humana Press",

number = "3",

}

TY - JOUR

T1 - Seizure detection in adult ICU patients based on changes in EEG synchronization likelihood

AU - Slooter, A J C

AU - Vriens, E M

AU - Leijten, F S S

AU - Spijkstra, J J

AU - Girbes, A R J

AU - van Huffelen, A C

AU - Stam, C J

PY - 2006

Y1 - 2006

N2 - INTRODUCTION: Seizures are common in Intensive Care Unit (ICU) patients, and may increase neuronal injury.PURPOSE: To explore the possible value of synchronization likelihood (SL) for the automatic detection of seizures in adult ICU patients.METHODS: We included EEGs from ICU patients with a variety of diagnoses. The gold standard for further analyses was the consensus judgment of three clinical neurophysiologists who classified 150 scalp EEG epochs as "definitely epileptiform," "definitely non epileptiform," or "uncertain." SL estimates the statistical interdependencies between two time series, such as two EEG channels. We computed the average synchronization by calculating the SL between one channel and every other channel, and taking the mean of these values.RESULTS: The mean SL in the 38 "definitely epileptiform" epochs ranged from 0.095 to 0.386 (mean 0.189; SD 0.066). In the 34 "definitely nonepileptiform" epochs the mean SL ranged from 0.087 to 0.158 (mean 0.115; SD 0.016; p < 0.0005). The area under the ROC curve was 0.812 (95% Confidence Interval 0.725 to 0.898).CONCLUSION: The mean SL may distinguish between seizure and nonseizure epochs, and may prove helpful to monitor epileptic activity in ICU patients.

AB - INTRODUCTION: Seizures are common in Intensive Care Unit (ICU) patients, and may increase neuronal injury.PURPOSE: To explore the possible value of synchronization likelihood (SL) for the automatic detection of seizures in adult ICU patients.METHODS: We included EEGs from ICU patients with a variety of diagnoses. The gold standard for further analyses was the consensus judgment of three clinical neurophysiologists who classified 150 scalp EEG epochs as "definitely epileptiform," "definitely non epileptiform," or "uncertain." SL estimates the statistical interdependencies between two time series, such as two EEG channels. We computed the average synchronization by calculating the SL between one channel and every other channel, and taking the mean of these values.RESULTS: The mean SL in the 38 "definitely epileptiform" epochs ranged from 0.095 to 0.386 (mean 0.189; SD 0.066). In the 34 "definitely nonepileptiform" epochs the mean SL ranged from 0.087 to 0.158 (mean 0.115; SD 0.016; p < 0.0005). The area under the ROC curve was 0.812 (95% Confidence Interval 0.725 to 0.898).CONCLUSION: The mean SL may distinguish between seizure and nonseizure epochs, and may prove helpful to monitor epileptic activity in ICU patients.

KW - Cerebral Cortex/physiopathology

KW - Cortical Synchronization

KW - Diagnosis, Differential

KW - Electroencephalography/statistics & numerical data

KW - Epilepsy/diagnosis

KW - Evoked Potentials/physiology

KW - Humans

KW - Intensive Care Units

KW - Likelihood Functions

KW - Nonlinear Dynamics

KW - Observer Variation

KW - ROC Curve

KW - Signal Processing, Computer-Assisted

KW - Status Epilepticus/diagnosis

U2 - https://doi.org/10.1385/NCC:5:3:186

DO - https://doi.org/10.1385/NCC:5:3:186

M3 - Article

C2 - 17290086

SN - 1541-6933

VL - 5

SP - 186

EP - 192

JO - Neurocritical Care

JF - Neurocritical Care

IS - 3

ER -