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

Imagining and building wise machines: the centrality of AI metacognition.

Trends in cognitive sciences·2026
Same author

Learning Compact Discriminant Representation via Low-Rank Bilinear Pooling.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

Prospector Heads: Generalized Feature Attribution for Large Models & Data.

ArXiv·2024
Same author

UncertaintyFuseNet: Robust uncertainty-aware hierarchical feature fusion model with Ensemble Monte Carlo Dropout for COVID-19 detection.

An international journal on information fusion·2022
Same author

Human activity recognition using tools of convolutional neural networks: A state of the art review, data sets, challenges, and future prospects.

Computers in biology and medicine·2022
Same author

Spatial Attention-Based 3D Graph Convolutional Neural Network for Sign Language Recognition.

Sensors (Basel, Switzerland)·2022

Related Experiment Video

Updated: Jan 16, 2026

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality
10:14

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality

Published on: May 10, 2024

1.8K

Temporal convolutional transformer for EEG based motor imagery decoding.

Hamdi Altaheri1, Fakhri Karray2,3, Amir-Hossein Karimi2,4

  • 1Department of Electrical and Computer Engineering, University of Waterloo, Waterloo, ON, Canada. haltaheri@uwaterloo.ca.

Scientific Reports
|September 26, 2025
PubMed
Summary
This summary is machine-generated.

TCFormer, a novel temporal convolutional Transformer, significantly enhances brain-computer interface (BCI) performance for decoding motor imagery (MI) from EEG signals. This advancement improves rehabilitation and control applications by accurately translating imagined movements.

Keywords:
Brain signal decodingConvolutional neural networkElectroencephalography (EEG)Grouped query attentionMotor imagery classificationTemporal convolutional networkTransformers

More Related Videos

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

12.2K
Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

44.0K

Related Experiment Videos

Last Updated: Jan 16, 2026

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality
10:14

Author Spotlight: Enhancing Neurorehabilitation Through EEG, Motor Imagery, and Virtual Reality

Published on: May 10, 2024

1.8K
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

12.2K
Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

44.0K

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Artificial Intelligence

Background:

  • Brain-computer interfaces (BCIs) leverage electroencephalography (EEG) to translate neural signals into commands.
  • Accurate decoding of motor imagery (MI) is crucial for effective BCI applications in rehabilitation and control.
  • Current BCI methods face challenges in capturing complex spatial-temporal dynamics in EEG signals.

Purpose of the Study:

  • To introduce TCFormer, a novel deep learning architecture for improved EEG-based motor imagery decoding.
  • To enhance the performance of BCIs by effectively processing complex EEG data.
  • To advance the capabilities of BCIs for rehabilitation, communication, and control.

Main Methods:

  • TCFormer integrates a multi-kernel convolutional neural network (MK-CNN) for spatial-temporal feature extraction.
  • A Transformer encoder with grouped query attention captures global contextual dependencies.
  • A temporal convolutional network (TCN) head with dilated causal convolutions learns long-range temporal patterns.

Main Results:

  • TCFormer achieved high average accuracies on benchmark datasets: 84.79% (BCIC IV-2a), 87.71% (BCIC IV-2b), and 96.27% (HGD).
  • The proposed architecture outperformed existing methods in motor imagery decoding.
  • The model demonstrated effectiveness in addressing the complexity of EEG signals.

Conclusions:

  • The TCFormer architecture significantly improves the accuracy of motor imagery decoding from EEG.
  • The integrated MK-CNN and Transformer design effectively captures spatial-temporal features and global dependencies.
  • TCFormer represents a promising advancement for practical BCI applications.