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

    991 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

    Fingerprint

    Dive into the research topics of 'Automated Algorithm for Generalized Tonic–Clonic Epileptic Seizure Onset Detection Based on sEMG Zero-Crossing Rate'. Together they form a unique fingerprint.

    Cite this