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

Advances in the Application of Brain-Computer Interface-Based Neurofeedback Training in the Rehabilitation of Patients with Major Depressive Disorder.

Neuropsychiatric disease and treatment·2026
Same author

Association of glycolipid metabolism 6 factors index with cardiovascular and metabolic diseases, and its predictive value in disease progression.

BMC endocrine disorders·2026
Same author

EEG-EMG spatiotemporal cross-attention fusion network for functional upper-limb movement classification.

Biomedical physics & engineering express·2026
Same author

Hemoglobin concentration sensing via a high-Q-factor Q-BIC-excited heterogeneous dielectric terahertz metasurface biosensor.

iScience·2026
Same author

[Reproducibility of lesion stiffness measured by shear wave elastography in Hashimoto's thyroiditis combined with thyroid cancer and its correlation with pathological features].

Lin chuang er bi yan hou tou jing wai ke za zhi = Journal of clinical otorhinolaryngology head and neck surgery·2026
Same author

Unraveling the significance of decorin in endometriosis development through single cell sequencing and experimental approaches.

PloS one·2026

Related Experiment Video

Updated: Jul 25, 2025

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

1.0K

MSATNet: multi-scale adaptive transformer network for motor imagery classification.

Lingyan Hu1,2, Weijie Hong3, Lingyu Liu4

  • 1School of Information and Engineering, Nanchang University, Nanchang, Jiangxi, China.

Frontiers in Neuroscience
|June 30, 2023
PubMed
Summary
This summary is machine-generated.

A new multi-scale adaptive transformer network (MSATNet) improves motor imagery brain-computer interface (MI-BCI) accuracy. This advanced model enhances feature extraction and cross-subject performance for better control of wheelchairs and prosthetics.

Keywords:
electroencephalogrammotor imagery classificationmulti-scale convolutiontransfer learningtransformer

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

444
Non-Invasive Modulation and Robotic Mapping of Motor Cortex in the Developing Brain
08:26

Non-Invasive Modulation and Robotic Mapping of Motor Cortex in the Developing Brain

Published on: July 1, 2019

6.8K

Related Experiment Videos

Last Updated: Jul 25, 2025

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

1.0K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

444
Non-Invasive Modulation and Robotic Mapping of Motor Cortex in the Developing Brain
08:26

Non-Invasive Modulation and Robotic Mapping of Motor Cortex in the Developing Brain

Published on: July 1, 2019

6.8K

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Machine Learning

Background:

  • Motor imagery brain-computer interfaces (MI-BCI) enable device control via thought.
  • Current MI-BCI models struggle with effective feature extraction and cross-subject generalization.

Purpose of the Study:

  • To introduce a novel Multi-Scale Adaptive Transformer Network (MSATNet) for enhanced motor imagery classification.
  • To address limitations in feature extraction and cross-subject performance in existing MI-BCI systems.

Main Methods:

  • Developed a Multi-Scale Feature Extraction (MSFE) module for robust feature identification.
  • Implemented an Adaptive Temporal Transformer (ATT) module to capture complex temporal dependencies.
  • Integrated a Subject Adapter (SA) module for efficient transfer learning across subjects.

Main Results:

  • MSATNet achieved superior classification accuracies on the BCI Competition IV 2a and 2b datasets.
  • Within-subject accuracies reached 81.75% and 89.34%.
  • Cross-subject accuracies reached 81.33% and 86.23%, outperforming benchmark models.

Conclusions:

  • The proposed MSATNet significantly improves motor imagery classification performance.
  • The MSATNet architecture offers a promising solution for developing more accurate and adaptable MI-BCI systems.