Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Algorithm architectures for patient dependent seizure detection.

Scott B Wilson1

  • 1Persyst Development Corporation, 1060 Sandretto Drive, Suite E2, Prescott, AZ 86305, USA. scottw@eeg-persyst.com

Clinical Neurophysiology : Official Journal of the International Federation of Clinical Neurophysiology
|April 8, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Optical Flow Estimation Improves Automated Seizure Detection in Neonatal EEG.

Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society·2020
Same author

Seizure Detection: Interreader Agreement and Detection Algorithm Assessments Using a Large Dataset.

Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society·2020
Same author

Semiautomated Spike Detection Software Persyst 13 Is Noninferior to Human Readers When Calculating the Spike-Wave Index in Electrical Status Epilepticus in Sleep.

Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society·2018
Same author

Reply to "What it should mean for an algorithm to pass a statistical Turing test for detection of epileptiform discharges".

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2017
Same author

Spike detection: Inter-reader agreement and a statistical Turing test on a large data set.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2016
Same author

Spike detection algorithm performance and methods of acquiring expert opinion.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2010
Same journal

Modulating corticospinal excitability with transcranial ultrasound stimulation: meta-analytic evidence of online and offline effects.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
Same journal

Fast oscillations as useful biomarkers of the degree of epileptogenicity in each generalized epilepsy syndrome.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
Same journal

Neuropsychological, biological, and electrophysiological outcomes of gamma-tACS in MCI-AD: A case series.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
Same journal

Enhanced broad-band intermuscular coherence in myoclonus: a targeted characterization study.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
Same journal

Perturbing the vestibular cortex with transcranial oscillatory currents uncovers early postural alterations in Parkinson's disease.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
Same journal

F-waves in primary lateral sclerosis: a window into spinal motoneuron hyperexcitability.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2026
See all related articles

A new seizure detection algorithm, AutoLearn, performs as well as or better than existing methods with minimal user input. This patient-independent algorithm can identify unusual seizures missed by current detectors.

Area of Science:

  • Neurology
  • Biomedical Engineering
  • Machine Learning

Background:

  • Automated seizure detection is crucial for epilepsy monitoring.
  • Current patient-independent algorithms often lack optimal performance.
  • Developing algorithms that require minimal user input is a significant challenge.

Purpose of the Study:

  • To develop and evaluate an on-the-fly algorithm for seizure detection.
  • To determine if improved performance over patient-independent algorithms can be achieved with minimal user input.
  • To create an algorithm that requires only the identification of the first one or two seizures.

Main Methods:

  • The AutoLearn algorithm, utilizing a probabilistic neural network (PNN), was tested on 209 seizures from epilepsy monitoring unit (EMU) and ambulatory recordings.

Related Experiment Videos

  • A construction algorithm and Taguchi design of experiments (DoE) were used to compare various algorithm architectures and identify significant factors.
  • The method focused on architectures with a single PNN per channel and segmentation for activity ranges.

    • Main Results:

    • Preferred architectures involved a single PNN per channel and segmentation for activity ranges.
    • The top two architectures were insensitive to other tested factors.
    • Algorithm training time was under 1 second, with approximately 2 minutes needed to process an 8-hour record.

    Conclusions:

    • The final algorithm, requiring only the marking of the first seizure, matched or exceeded the performance of existing seizure detectors on EMU and ambulatory records.
    • The algorithm's performance approached that of human experts on EMU records.
    • This method can detect unusual seizures missed by current detectors and aids in developing adaptive, patient-independent seizure detection systems.