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 Video

Updated: Jan 9, 2026

Author Spotlight: Unraveling Seizure Dynamics and Novel Therapeutics for Status Epilepticus Using CMOS High-Density Microelectrode Array Systems
06:28

Author Spotlight: Unraveling Seizure Dynamics and Novel Therapeutics for Status Epilepticus Using CMOS High-Density Microelectrode Array Systems

Published on: September 27, 2024

3.2K

A Lightweight CNN-LSTM Approach for Accurate EEG Normal-Abnormal Classification.

Lan Wei, Catherine Mooney

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
    |December 3, 2025
    PubMed
    Summary

    This study introduces a lightweight model for classifying normal and abnormal electroencephalogram (EEG) signals using a single channel. The approach enhances diagnostic efficiency and interpretability for neurological condition identification.

    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

    Handwritten tax receipt recognition method based on improved DBNet and CRNN.

    PloS one·2026
    Same author

    Loss of the type VII secretion ATPase EssC promotes biofilm formation of <i>Staphylococcus</i> <i>aureus</i> under acidic stress.

    Biofilm·2026
    Same author

    The <i>FHY3</i>/<i>FAR1</i> Gene Family in Plants: Transposase-Derived Transcription Factors as Master Integrators of Light Signaling and Plant Development.

    Plants (Basel, Switzerland)·2026
    Same author

    Bacillus coagulans BC66 Attenuates Klebsiella pneumoniae-Induced Acute Lung Injury in Rabbits Associated With Modulation of Th17/Treg Immune Balance and Through the Keap1/Nrf2/HO-1 Signalling Pathway.

    Probiotics and antimicrobial proteins·2026
    Same author

    Proposing an explanatory framework based on the fear-avoidance model: a mixed-methods analysis of kinesiophobia in patients after percutaneous coronary intervention in home-based cardiac rehabilitation.

    Frontiers in cardiovascular medicine·2026
    Same author

    Comparisons of the Burden and Quality of Care of Nonrheumatic Valvular Heart Disease in Asia and Globally.

    Journal of the American Heart Association·2026

    Area of Science:

    • Neurology
    • Artificial Intelligence
    • Biomedical Engineering

    Background:

    • Electroencephalogram (EEG) is crucial for diagnosing neurological disorders, with classification being a primary diagnostic step.
    • Previous EEG classification methods often lack reproducibility due to reliance on non-public datasets.
    • There is a need for efficient and accessible methods for EEG analysis.

    Purpose of the Study:

    • To develop a lightweight Convolutional Neural Network-LSTM (CNN-LSTM) model for classifying normal versus abnormal EEG signals.
    • To utilize the publicly available TUH abnormal EEG dataset for training and validation.
    • To improve the efficiency and interpretability of EEG analysis.

    Main Methods:

    • A lightweight CNN-LSTM model was designed for binary classification of EEG signals.

    Related Experiment Videos

    Last Updated: Jan 9, 2026

    Author Spotlight: Unraveling Seizure Dynamics and Novel Therapeutics for Status Epilepticus Using CMOS High-Density Microelectrode Array Systems
    06:28

    Author Spotlight: Unraveling Seizure Dynamics and Novel Therapeutics for Status Epilepticus Using CMOS High-Density Microelectrode Array Systems

    Published on: September 27, 2024

    3.2K
  • The model was trained and evaluated using the single T5-O1 channel from the TUH abnormal EEG dataset.
  • Gradient-weighted Class Activation Mapping (Grad-CAM) was employed for model visualization and interpretability.

    • Main Results:

    • The proposed CNN-LSTM model achieved effective classification of normal and abnormal EEGs.
    • Using a single EEG channel (T5-O1) simplified the model and reduced computational requirements.
    • Automatic feature selection via convolutional layers enhanced model efficiency.

    Conclusions:

    • The developed lightweight CNN-LSTM model offers an efficient and reproducible method for EEG classification.
    • The single-channel approach reduces complexity and computational cost, making it more accessible.
    • Model interpretability was improved through visualization techniques, aiding in understanding diagnostic features.