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

ROI-guided virtual narrow band imaging for laryngeal cancer screening.

iScience·2026
Same author

TF-MCL: Time-frequency fusion and multi-domain cross-loss for self-supervised depression detection.

Biomedical physics & engineering express·2026
Same author

Penidelanones A-F: furancarboxylic acids from the endophytic fungus <i>Penicillium daleae</i>.

Journal of Asian natural products research·2026
Same author

TrkB promotes the neuronal secretion of soluble Siglec-2 (CD22) to mitigate microglial activation and alleviate depression-like behaviors in male mice.

Molecular psychiatry·2026
Same author

The protein RAB5IF promotes BDNF signaling by stimulating the SUMOylation of Gα<sub>i1/3</sub> to reduce depressive-like behaviors in mice.

Science signaling·2026
Same author

Aflatoxin and Liver Cancer in China: The Evolving Research Landscape.

Toxins·2026

Related Experiment Video

Updated: Sep 10, 2025

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
05:19

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment

Published on: July 7, 2023

2.5K

Major depressive disorder detection via temporal-frequency-spatial transformer with sub-domain knowledge alignment

Chen-Yang Xu1, Fei-Yi Fan2, Li-Xuan Zhao1

  • 1School of Electrical and Information Engineering, Tianjin University, 300072, Tianjin, China.

Neural Networks : the Official Journal of the International Neural Network Society
|August 27, 2025
PubMed
Summary

This study introduces a new AI model for detecting Major Depressive Disorder (MDD) using electroencephalogram (EEG) data. The TFST-SDKA model enhances accuracy by focusing on frequency features and sub-domain alignment for better generalization.

Keywords:
Frequency attention (FA)MDD detectionSub-domain knowledge alignment (SDKA)Temporal-frequency-spatial transformer (TFST)

More Related Videos

Automatic Detection of Highly Organized Theta Oscillations in the Murine EEG
09:35

Automatic Detection of Highly Organized Theta Oscillations in the Murine EEG

Published on: March 10, 2017

9.3K
Concurrent EEG and Functional MRI Recording and Integration Analysis for Dynamic Cortical Activity Imaging
11:28

Concurrent EEG and Functional MRI Recording and Integration Analysis for Dynamic Cortical Activity Imaging

Published on: June 30, 2018

11.8K

Related Experiment Videos

Last Updated: Sep 10, 2025

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
05:19

Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment

Published on: July 7, 2023

2.5K
Automatic Detection of Highly Organized Theta Oscillations in the Murine EEG
09:35

Automatic Detection of Highly Organized Theta Oscillations in the Murine EEG

Published on: March 10, 2017

9.3K
Concurrent EEG and Functional MRI Recording and Integration Analysis for Dynamic Cortical Activity Imaging
11:28

Concurrent EEG and Functional MRI Recording and Integration Analysis for Dynamic Cortical Activity Imaging

Published on: June 30, 2018

11.8K

Area of Science:

  • Neuroscience
  • Artificial Intelligence
  • Biomedical Engineering

Background:

  • Major Depressive Disorder (MDD) significantly impacts patient health, necessitating accurate detection methods.
  • Current electroencephalogram (EEG) based detection methods face challenges due to inter-subject variability, limiting cross-subject accuracy.
  • Existing Transformer models for MDD detection often overlook crucial frequency features and sub-domain alignment in domain adaptation.

Purpose of the Study:

  • To improve the accuracy and generalization of Transformer-based MDD detection using EEG data.
  • To address the limitations of global domain adaptation by incorporating fine-grained frequency features and sub-domain alignment.
  • To develop a novel model that enhances EEG feature representation and aligns source and target domain features effectively.

Main Methods:

  • Proposed the Temporal-Frequency-Spatial Transformer (TFST) integrated with Sub-Domain Knowledge Alignment (SDKA) for MDD detection.
  • Implemented SDKA to classify subjects into sub-domains, extracting fine-grained discriminative information to improve generalization.
  • Introduced a Frequency Attention (FA) mechanism using Discrete Cosine Transform (DCT) to capture and combine multiple frequency information from EEG signals.

Main Results:

  • The TFST-SDKA model demonstrated superior performance compared to state-of-the-art methods on the MODMA and PRED+CT datasets.
  • Achieved accuracy improvements of 1.42% on the MODMA dataset and 1.16% on the PRED+CT dataset.
  • The model effectively enhanced EEG feature representation and improved source-target domain alignment.

Conclusions:

  • The proposed TFST-SDKA model offers a significant advancement in EEG-based MDD detection.
  • Incorporating fine-grained frequency features and sub-domain alignment is crucial for enhancing model generalization.
  • The developed model shows strong potential for accurate and reliable clinical application in MDD diagnosis.