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

Data-Efficient Bone Segmentation Using Feature Pyramid- Based SegFormer.

Sensors (Basel, Switzerland)·2025
Same author

Attention-Based PSO-LSTM for Emotion Estimation Using EEG.

Sensors (Basel, Switzerland)·2025
See all related articles

Related Experiment Video

Updated: Feb 28, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.5K

EEG-Based Emotion Estimation Model Integrating Structural and Time-Series Information Based on Deep Learning

Kota Tsuji1, Keiko Ono2, Takuya Futagami2

  • 1Graduate School of Science and Engineering Information and Computer Science, Doshisha University, Kyoto 610-0394, Japan.

Sensors (Basel, Switzerland)
|February 27, 2026
PubMed
Summary

This study introduces an automated EEG emotion recognition framework using Graph Convolutional Networks and LSTMs. It enhances accuracy and adaptability by optimizing neural network architectures for individual brain patterns.

Keywords:
DARTSDEAPEEGGCNLSTMemotion estimationfour-class classification

More Related Videos

Investigating the Function of Deep Cortical and Subcortical Structures Using Stereotactic Electroencephalography: Lessons from the Anterior Cingulate Cortex
09:00

Investigating the Function of Deep Cortical and Subcortical Structures Using Stereotactic Electroencephalography: Lessons from the Anterior Cingulate Cortex

Published on: April 15, 2015

12.9K
Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

767

Related Experiment Videos

Last Updated: Feb 28, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.5K
Investigating the Function of Deep Cortical and Subcortical Structures Using Stereotactic Electroencephalography: Lessons from the Anterior Cingulate Cortex
09:00

Investigating the Function of Deep Cortical and Subcortical Structures Using Stereotactic Electroencephalography: Lessons from the Anterior Cingulate Cortex

Published on: April 15, 2015

12.9K
Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

767

Area of Science:

  • Neuroscience
  • Machine Learning
  • Biomedical Engineering

Background:

  • Emotion recognition is vital for mental health and marketing.
  • Facial and vocal cues are unreliable; Electroencephalography (EEG) offers robust neural data.
  • Current EEG models (CNNs, LSTMs) have limitations in connectivity, variability, and manual design.

Purpose of the Study:

  • To develop an automated, adaptive EEG-based emotion recognition system.
  • To overcome limitations of handcrafted models and manual hyperparameter tuning.
  • To improve scalability and personalization in emotion recognition.

Main Methods:

  • Proposed a dual-pipeline architecture integrating frequency-domain (Graph Convolutional Network - GCN) and time-domain (LSTM with Channel Attention) EEG features.
  • Employed Differentiable Architecture Search (DARTS) for automated, individualized architecture optimization.
  • Modeled electrode connectivity using GCN and emphasized subject-specific channels with attention.

Main Results:

  • The framework achieved competitive accuracy and high adaptability compared to existing methods.
  • Demonstrated the effectiveness of integrating GCN, LSTM, channel attention, and architecture search.
  • Showcased significant reduction in search cost via automated architecture discovery.

Conclusions:

  • The proposed method offers a novel, automated approach to EEG emotion recognition.
  • Achieved superior performance and adaptability by optimizing architectures for individual neural patterns.
  • Paved the way for more personalized and scalable emotion recognition applications.