Seizure Onset Detection based on one sEMG channel

Isa Conradsen; Sandor Beniczky; Karsten Hoppe; Peter Wolf; Thomas Sams; Helge Bjarup Dissing Sørensen

doi:10.1109/IEMBS.2011.6091901

Seizure Onset Detection based on one sEMG channel

Isa Conradsen, Sandor Beniczky, Karsten Hoppe, Peter Wolf, Thomas Sams, Helge Bjarup Dissing Sørensen

Department of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

We present a new method to detect seizure onsets of tonic-clonic epileptic seizures based on surface electromyography (sEMG) data. The proposed method is generic and based on a single channel making it ideal for a small detection or monitoring device. The sEMG signal is high-pass filtered with a Butterworth filter with a cut-off frequency of 150 Hz. The number of zero-crossings with a hysteresis of ±50μV is the only feature extracted. The number of counts in a window of 1 second and the number of windows to make a detection is tested with a leave-one-out method. On 6 patients the method performs with a sensitivity of 100%, a median latency of 7.6 seconds and a median false detection rate of 0.04/h.

Original language	English
Title of host publication	Proceedings of the 33rd Annual International Conference of the IEEE EMBS
Publisher	IEEE
Publication date	2011
Pages	7715-7718
ISBN (Print)	978-1-4244-4122-8
DOIs	https://doi.org/10.1109/IEMBS.2011.6091901
Publication status	Published - 2011
Event	33rd Annual International Conference of the IEEE EMBS - Boston, Massachusetts USA Duration: 1 Jan 2011 → …

Conference

Conference	33rd Annual International Conference of the IEEE EMBS
City	Boston, Massachusetts USA
Period	01/01/2011 → …

Keywords

Sensitivity
Conferences
Training
Epilepsy
Muscles
Feature extraction
Monitoring

Access to Document

10.1109/IEMBS.2011.6091901

Fingerprint Dive into the research topics of 'Seizure Onset Detection based on one sEMG channel'. Together they form a unique fingerprint.

Cite this

@inproceedings{2a8c0602eaab4176bc609e8db1109e89,

title = "Seizure Onset Detection based on one sEMG channel",

abstract = "We present a new method to detect seizure onsets of tonic-clonic epileptic seizures based on surface electromyography (sEMG) data. The proposed method is generic and based on a single channel making it ideal for a small detection or monitoring device. The sEMG signal is high-pass filtered with a Butterworth filter with a cut-off frequency of 150 Hz. The number of zero-crossings with a hysteresis of ±50μV is the only feature extracted. The number of counts in a window of 1 second and the number of windows to make a detection is tested with a leave-one-out method. On 6 patients the method performs with a sensitivity of 100%, a median latency of 7.6 seconds and a median false detection rate of 0.04/h.",

keywords = "Sensitivity, Conferences, Training, Epilepsy, Muscles, Feature extraction, Monitoring",

author = "Isa Conradsen and Sandor Beniczky and Karsten Hoppe and Peter Wolf and Thomas Sams and S{\o}rensen, {Helge Bjarup Dissing}",

year = "2011",

doi = "10.1109/IEMBS.2011.6091901",

language = "English",

isbn = "978-1-4244-4122-8",

pages = "7715--7718",

booktitle = "Proceedings of the 33rd Annual International Conference of the IEEE EMBS",

publisher = "IEEE",

address = "United States",

note = "33rd Annual International Conference of the IEEE EMBS ; Conference date: 01-01-2011",

}

Seizure Onset Detection based on one sEMG channel. / Conradsen, Isa; Beniczky, Sandor; Hoppe, Karsten; Wolf, Peter; Sams, Thomas; Sørensen, Helge Bjarup Dissing.

Proceedings of the 33rd Annual International Conference of the IEEE EMBS. IEEE, 2011. p. 7715-7718.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Seizure Onset Detection based on one sEMG channel

AU - Conradsen, Isa

AU - Beniczky, Sandor

AU - Hoppe, Karsten

AU - Wolf, Peter

AU - Sams, Thomas

AU - Sørensen, Helge Bjarup Dissing

PY - 2011

Y1 - 2011

N2 - We present a new method to detect seizure onsets of tonic-clonic epileptic seizures based on surface electromyography (sEMG) data. The proposed method is generic and based on a single channel making it ideal for a small detection or monitoring device. The sEMG signal is high-pass filtered with a Butterworth filter with a cut-off frequency of 150 Hz. The number of zero-crossings with a hysteresis of ±50μV is the only feature extracted. The number of counts in a window of 1 second and the number of windows to make a detection is tested with a leave-one-out method. On 6 patients the method performs with a sensitivity of 100%, a median latency of 7.6 seconds and a median false detection rate of 0.04/h.

AB - We present a new method to detect seizure onsets of tonic-clonic epileptic seizures based on surface electromyography (sEMG) data. The proposed method is generic and based on a single channel making it ideal for a small detection or monitoring device. The sEMG signal is high-pass filtered with a Butterworth filter with a cut-off frequency of 150 Hz. The number of zero-crossings with a hysteresis of ±50μV is the only feature extracted. The number of counts in a window of 1 second and the number of windows to make a detection is tested with a leave-one-out method. On 6 patients the method performs with a sensitivity of 100%, a median latency of 7.6 seconds and a median false detection rate of 0.04/h.

KW - Sensitivity

KW - Conferences

KW - Training

KW - Epilepsy

KW - Muscles

KW - Feature extraction

KW - Monitoring

U2 - 10.1109/IEMBS.2011.6091901

DO - 10.1109/IEMBS.2011.6091901

M3 - Article in proceedings

C2 - 22256126

SN - 978-1-4244-4122-8

SP - 7715

EP - 7718

BT - Proceedings of the 33rd Annual International Conference of the IEEE EMBS

PB - IEEE

T2 - 33rd Annual International Conference of the IEEE EMBS

Y2 - 1 January 2011

ER -