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

A stochastic framework for evaluating seizure prediction algorithms using hidden Markov models.

Stephen Wong1, Andrew B Gardner, Abba M Krieger

  • 1Department of Neurology, 2 Ravdin Penn Epilepsy Center, Hospital of the University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104, USA. swong@swong.org

Journal of Neurophysiology
|October 6, 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

Virtual Responsive Neurostimulation Implantation: From Intracranial Connectivity to Optimized Lead Placement.

medRxiv : the preprint server for health sciences·2026
Same author

Cortical stimulation reveals effective disconnection of the epileptogenic network at seizure onset.

Epilepsia·2026
Same author

Association between Interictal Spike Rate and Seizure Frequency in a Large Epilepsy Cohort.

medRxiv : the preprint server for health sciences·2026
Same author

The seizure embedding map: A spatio-temporal transformer for comparing patients by ictal intracranial EEG features at scale.

Journal of neural engineering·2026
Same author

Identifying High-Risk Stage I NSCLC: The Promises and Pitfalls of Tumor-Informed ctDNA.

Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer·2026
Same author

Regional excitability, not epileptic pathology, drives stimulation-evoked interictal spike increases.

medRxiv : the preprint server for health sciences·2026
Same journal

Comprehensive Analysis of Auditory Nerve Fiber Responses using Fiber-Specific Modeling.

Journal of neurophysiology·2026
Same journal

HCN channels modulate the medium afterhyperpolarization and adjust the firing gain of fast alpha motoneurons in mice.

Journal of neurophysiology·2026
Same journal

Targeting intracranial electrical stimulation to network regions defined within individuals causes network-level effects.

Journal of neurophysiology·2026
Same journal

When "Noise" Isn't Simply Noise: Deterministic Postural Drive During Noisy Galvanic Vestibular Stimulation (nGVS).

Journal of neurophysiology·2026
Same journal

Abrupt Scene Onsets and Gradually Emerging Scene Information Produce Distinct EEG Decoding Dynamics.

Journal of neurophysiology·2026
Same journal

From discovery to translation: charting a course for the <i>Journal of Neurophysiology</i>.

Journal of neurophysiology·2026
See all related articles

Developing a new statistical framework for epilepsy seizure prediction is crucial for improving implantable device therapy. This hidden Markov model (HMM) approach offers a standardized method for validating seizure prediction algorithms.

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Computational Statistics

Background:

  • Implantable devices for epilepsy treatment are in clinical trials, with potential for improved efficacy through pre-seizure onset therapy.
  • Prospective seizure prediction remains a challenge due to the lack of a standardized statistical framework for modeling seizure generation and validating algorithms.

Purpose of the Study:

  • To introduce a novel stochastic framework for epilepsy seizure prediction using a three-state hidden Markov model (HMM).
  • To provide a method for validating seizure prediction algorithm performance and facilitating collaborative research.

Main Methods:

  • Developed a three-state hidden Markov model (HMM) representing interictal, preictal, and seizure states, allowing transitions back to the interictal state.
  • Derived equations for type I and type II errors based on seizure frequency, interictal data duration, and prediction horizon.

Related Experiment Videos

  • Validated the framework using a novel seizure detection algorithm that demonstrated seizure onset prediction.

    • Main Results:

    • The proposed HMM framework accommodates clipped EEG data and formalizes statistical validation concepts.
    • The model's utility was demonstrated through a novel algorithm that successfully predicted seizure onset.
    • Derived equations provide a quantitative basis for evaluating prediction algorithm performance.

    Conclusions:

    • The novel HMM framework offers a vital tool for the design and validation of epilepsy seizure prediction algorithms.
    • This standardized approach is expected to enhance the interpretation of existing studies and foster collaborative research in the field.
    • The framework's ability to model transitions back to the interictal state reflects clinical observations and improves model interpretability.