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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

sEEG-Suite: An Interactive Pipeline for Semi-Automated Contact Localization and Anatomical Labeling with Brainstorm.

bioRxiv : the preprint server for biology·2025
Same author

UrBAN: Urban Beehive Acoustics and PheNotyping Dataset.

Scientific data·2025
Same author

One hundred years of EEG for brain and behaviour research.

Nature human behaviour·2024
Same author

MSPB: a longitudinal multi-sensor dataset with phenotypic trait measurements from honey bees.

Scientific data·2024
Same author

EEG Amplitude Modulation Analysis across Mental Tasks: Towards Improved Active BCIs.

Sensors (Basel, Switzerland)·2023
Same author

Neuromorphic Computing via Fission-based Broadband Frequency Generation.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2023

Related Experiment Video

Updated: Aug 9, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.2K

Using CNN Saliency Maps and EEG Modulation Spectra for Improved and More Interpretable Machine Learning-Based

Marilia Lopes1, Raymundo Cassani2, Tiago H Falk1

  • 1Institute National de la Recherche Scientifique (INRS-EMT), University of Quebec, Montreal, Canada.

Computational Intelligence and Neuroscience
|February 23, 2023
PubMed
Summary

Deep learning enhances Alzheimer's disease (AD) diagnostics using electroencephalography (EEG) power modulation spectrograms. New data-driven biomarkers outperform existing methods and are not confounded by age, offering improved automated AD detection.

More Related Videos

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
08:51

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

5.7K
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

15.7K

Related Experiment Videos

Last Updated: Aug 9, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.2K
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
08:51

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

5.7K
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

15.7K

Area of Science:

  • Neuroscience
  • Medical Technology
  • Artificial Intelligence

Background:

  • Resting-state electroencephalography (EEG) biomarkers show promise for Alzheimer's disease (AD) diagnosis.
  • Current state-of-the-art methods rely on visually identified regions in power modulation spectrograms.

Purpose of the Study:

  • To develop and validate novel Alzheimer's disease (AD) biomarkers using deep learning on EEG power modulation spectrograms.
  • To identify optimal diagnostic regions in a data-driven manner using convolutional neural networks (CNNs) and saliency maps.

Main Methods:

  • Convolutional neural networks (CNNs) combined with saliency maps were trained on EEG power modulation spectrograms.
  • Experiments included 54 participants (20 controls, 19 mild AD, 15 moderate-to-severe AD).
  • Five classification tasks were performed: three-class, early-stage, and severity detection.

Main Results:

  • The proposed deep learning biomarkers significantly outperformed the state-of-the-art benchmark across all five classification tasks.
  • Novel, complementary diagnostic regions within modulation spectrograms were identified.
  • The developed biomarkers demonstrated no significant age-related confounding, enhancing their diagnostic utility.

Conclusions:

  • Deep learning-based biomarkers derived from EEG power modulation spectrograms offer superior performance for Alzheimer's disease (AD) diagnostics.
  • The data-driven approach identifies effective diagnostic regions beyond visual inspection.
  • These age-independent biomarkers represent a significant advancement in automated AD detection and characterization.