Automated Algorithm for Generalized Tonic–Clonic Epileptic Seizure Onset Detection Based on sEMG Zero-Crossing Rate

Isa Conradsen, Sándor Beniczky, Karsten Hoppe, Peter Wolf, Helge Bjarup Dissing Sørensen

Research output: Contribution to journalJournal articleResearchpeer-review

485 Downloads (Pure)

Abstract

Patients are not able to call for help during a generalized tonic–clonic epileptic seizure. Our objective was to develop a robust generic algorithm for automatic detection of tonic–clonic seizures, based on surface electromyography (sEMG) signals suitable for a portable device. Twenty-two seizures were analyzed from 11 consecutive patients. Our method is based on a high-pass filtering with a cutoff at 150 Hz, and monitoring a count of zero crossings with a hysteresis of $\pm 50\,\mu \hbox{V}$ . Based on data from one sEMG electrode (on the deltoid muscle), we achieved a sensitivity of 100% with a mean detection latency of 13.7 s, while the rate of false detection was limited to 1 false alarm per 24 h. The overall performance of the presented generic algorithm is adequate for clinical implementation.
Original languageEnglish
JournalI E E E Transactions on Biomedical Engineering
Volume59
Issue number2
Pages (from-to)579-585
ISSN0018-9294
DOIs
Publication statusPublished - 2012

Keywords

  • Tonic–clonic
  • Epilepsy
  • Surface electromyography (sEMG)
  • Seizure detection

Cite this

@article{4f1a8e15aebe40aebff21e52ee17b72d,
title = "Automated Algorithm for Generalized Tonic–Clonic Epileptic Seizure Onset Detection Based on sEMG Zero-Crossing Rate",
abstract = "Patients are not able to call for help during a generalized tonic–clonic epileptic seizure. Our objective was to develop a robust generic algorithm for automatic detection of tonic–clonic seizures, based on surface electromyography (sEMG) signals suitable for a portable device. Twenty-two seizures were analyzed from 11 consecutive patients. Our method is based on a high-pass filtering with a cutoff at 150 Hz, and monitoring a count of zero crossings with a hysteresis of $\pm 50\,\mu \hbox{V}$ . Based on data from one sEMG electrode (on the deltoid muscle), we achieved a sensitivity of 100{\%} with a mean detection latency of 13.7 s, while the rate of false detection was limited to 1 false alarm per 24 h. The overall performance of the presented generic algorithm is adequate for clinical implementation.",
keywords = "Tonic–clonic, Epilepsy, Surface electromyography (sEMG), Seizure detection",
author = "Isa Conradsen and S{\'a}ndor Beniczky and Karsten Hoppe and Peter Wolf and S{\o}rensen, {Helge Bjarup Dissing}",
year = "2012",
doi = "10.1109/TBME.2011.2178094",
language = "English",
volume = "59",
pages = "579--585",
journal = "I E E E Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "Institute of Electrical and Electronics Engineers",
number = "2",

}

Automated Algorithm for Generalized Tonic–Clonic Epileptic Seizure Onset Detection Based on sEMG Zero-Crossing Rate. / Conradsen, Isa; Beniczky, Sándor; Hoppe, Karsten; Wolf, Peter; Sørensen, Helge Bjarup Dissing.

In: I E E E Transactions on Biomedical Engineering, Vol. 59, No. 2, 2012, p. 579-585.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Automated Algorithm for Generalized Tonic–Clonic Epileptic Seizure Onset Detection Based on sEMG Zero-Crossing Rate

AU - Conradsen, Isa

AU - Beniczky, Sándor

AU - Hoppe, Karsten

AU - Wolf, Peter

AU - Sørensen, Helge Bjarup Dissing

PY - 2012

Y1 - 2012

N2 - Patients are not able to call for help during a generalized tonic–clonic epileptic seizure. Our objective was to develop a robust generic algorithm for automatic detection of tonic–clonic seizures, based on surface electromyography (sEMG) signals suitable for a portable device. Twenty-two seizures were analyzed from 11 consecutive patients. Our method is based on a high-pass filtering with a cutoff at 150 Hz, and monitoring a count of zero crossings with a hysteresis of $\pm 50\,\mu \hbox{V}$ . Based on data from one sEMG electrode (on the deltoid muscle), we achieved a sensitivity of 100% with a mean detection latency of 13.7 s, while the rate of false detection was limited to 1 false alarm per 24 h. The overall performance of the presented generic algorithm is adequate for clinical implementation.

AB - Patients are not able to call for help during a generalized tonic–clonic epileptic seizure. Our objective was to develop a robust generic algorithm for automatic detection of tonic–clonic seizures, based on surface electromyography (sEMG) signals suitable for a portable device. Twenty-two seizures were analyzed from 11 consecutive patients. Our method is based on a high-pass filtering with a cutoff at 150 Hz, and monitoring a count of zero crossings with a hysteresis of $\pm 50\,\mu \hbox{V}$ . Based on data from one sEMG electrode (on the deltoid muscle), we achieved a sensitivity of 100% with a mean detection latency of 13.7 s, while the rate of false detection was limited to 1 false alarm per 24 h. The overall performance of the presented generic algorithm is adequate for clinical implementation.

KW - Tonic–clonic

KW - Epilepsy

KW - Surface electromyography (sEMG)

KW - Seizure detection

U2 - 10.1109/TBME.2011.2178094

DO - 10.1109/TBME.2011.2178094

M3 - Journal article

C2 - 22156944

VL - 59

SP - 579

EP - 585

JO - I E E E Transactions on Biomedical Engineering

JF - I E E E Transactions on Biomedical Engineering

SN - 0018-9294

IS - 2

ER -