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 B D Sorensen

doi:10.1109/TBME.2011.2178094

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 B D Sorensen

Filadelfia

Publikation: Bidrag til tidsskrift › Artikel › Forskning › peer review

Abstrakt

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 ±50 μV . 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.

Originalsprog	Engelsk
Sider (fra-til)	579-85
Antal sider	7
Tidsskrift	IEEE Transactions on Biomedical Engineering
Vol/bind	59
Udgave nummer	2
DOI	https://doi.org/10.1109/TBME.2011.2178094
Status	Udgivet - feb. 2012

Adgang til dokumentet

10.1109/TBME.2011.2178094

Fingeraftryk Udforsk hvilke forskningsemner 'Automated algorithm for generalized tonic-clonic epileptic seizure onset detection based on sEMG zero-crossing rate' indeholder.

Citationsformater

@article{9aa67d46b5594e4ab7308da192ca9936,

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 ±50 μV . 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 = "Adult, Algorithms, Child, Diagnosis, Computer-Assisted/methods, Electromyography/methods, Epilepsy, Tonic-Clonic/diagnosis, Female, Humans, Male, Middle Aged, Seizures/diagnosis, Sensitivity and Specificity, Signal Processing, Computer-Assisted",

author = "Isa Conradsen and S{\'a}ndor Beniczky and Karsten Hoppe and Peter Wolf and Sorensen, {Helge B D}",

note = "{\textcopyright} 2011 IEEE",

year = "2012",

month = feb,

doi = "10.1109/TBME.2011.2178094",

language = "English",

volume = "59",

pages = "579--85",

journal = "IEEE Transactions on Biomedical Engineering",

issn = "0018-9294",

publisher = "IEEE Computer Society",

number = "2",

}

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 - Sorensen, Helge B D

PY - 2012/2

Y1 - 2012/2

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 ±50 μV . 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 ±50 μV . 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 - Adult

KW - Algorithms

KW - Child

KW - Diagnosis, Computer-Assisted/methods

KW - Electromyography/methods

KW - Epilepsy, Tonic-Clonic/diagnosis

KW - Female

KW - Humans

KW - Male

KW - Middle Aged

KW - Seizures/diagnosis

KW - Sensitivity and Specificity

KW - Signal Processing, Computer-Assisted

U2 - 10.1109/TBME.2011.2178094

DO - 10.1109/TBME.2011.2178094

M3 - Article

C2 - 22156944

VL - 59

SP - 579

EP - 585

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 2

ER -