Seizure Onset Detection based on one sEMG channel

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

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


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 languageEnglish
Title of host publicationProceedings of the 33rd Annual International Conference of the IEEE EMBS
Publication date2011
ISBN (Print)978-1-4244-4122-8
Publication statusPublished - 2011
Event33rd Annual International Conference of the IEEE EMBS - Boston, Massachusetts USA
Duration: 1 Jan 2011 → …


Conference33rd Annual International Conference of the IEEE EMBS
CityBoston, Massachusetts USA
Period01/01/2011 → …


  • Sensitivity
  • Conferences
  • Training
  • Epilepsy
  • Muscles
  • Feature extraction
  • Monitoring

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

Cite this