Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Effects of non-invasive brain stimulation on spontaneous neural activity in post-stroke aphasia: a protocol for a coordinate-based meta-analysis of fALFF studies.

Frontiers in neurology·2026

Same author

A novel phase-difference transcranial alternating current stimulation system enables precise dual-site neuromodulation.

Frontiers in neuroscience·2026

Same author

Non-invasive CT-based Deep Learning for Human Papillomavirus Status Prediction in Oropharyngeal Cancer.

Academic radiology·2026

Same author

Brain-Computer Interface Combined with Functional Electrical Stimulation for Post-Stroke Upper Limb Motor Recovery: A Systematic Review and Meta-Analysis.

Clinical EEG and neuroscience·2026

Same author

The mechanism of external cueing interventions in improving freezing of gait in Parkinson's disease: an fNIRS study.

Frontiers in aging neuroscience·2026

Same author

Global trends in rehabilitation needs for mental disorders in 204 countries and territories, 1990-2021: a cross-sectional analysis of the global burden of disease study 2021.

European archives of psychiatry and clinical neuroscience·2026

Search research articles

Related Experiment Video

Updated: Jun 7, 2025

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

Partial prior transfer learning based on self-attention CNN for EEG decoding in stroke patients.

Jun Ma¹, Wanlu Ma², Jingjing Zhang³

¹Department of Rehabilitation Medicine, Tong Ren Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, 200336, China. mj1990@shu.edu.cn.

Scientific Reports

|November 15, 2024

Summary

This study introduces a novel Self-Attention Convolutional Neural Network with Partial Prior Transfer Learning (SACNN-PPTL) to enhance motor imagery brain-computer interface (MI-BCI) decoding for stroke rehabilitation. The new method significantly improves classification accuracy for complex upper limb tasks.

Keywords:

Convolutional neural network EEG decoding Motor imagery Self-attention Transfer learning

More Related Videos

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

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

Published on: July 26, 2013

Simultaneous Eye Tracking and Single-Neuron Recordings in Human Epilepsy Patients

Simultaneous Eye Tracking and Single-Neuron Recordings in Human Epilepsy Patients

Published on: June 17, 2019

Related Experiment Videos

Last Updated: Jun 7, 2025

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

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

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

Published on: July 26, 2013

Simultaneous Eye Tracking and Single-Neuron Recordings in Human Epilepsy Patients

Simultaneous Eye Tracking and Single-Neuron Recordings in Human Epilepsy Patients

Published on: June 17, 2019

Area of Science:

Neuroscience
Biomedical Engineering
Artificial Intelligence

Background:

Motor imagery-based brain-computer interfaces (MI-BCI) aid stroke patients in activating brain motor regions.
Multi-task upper limb MI is beneficial for rehabilitation but poses decoding challenges.

Purpose of the Study:

To propose a novel method, SACNN-PPTL, for improving MI-BCI classification performance in stroke patients.
To enhance the decoding of complex, multi-task motor imagery for upper limb rehabilitation.

Main Methods:

Developed a Self-Attention Convolutional Neural Network (SACNN) incorporating temporal, spatial, and feature generalization modules.
Implemented Partial Prior Transfer Learning (PPTL) to balance model generalization and target domain specificity.
Evaluated SACNN-PPTL against five backbone networks and three training modes.

Main Results:

SACNN-PPTL achieved a classification accuracy of 55.4%±0.17 for four types of MI tasks in 22 stroke patients.
This accuracy was significantly higher than all comparison algorithms (P < 0.05).

Conclusions:

SACNN-PPTL effectively enhances the decoding performance of MI tasks for stroke rehabilitation.
The proposed method shows promise for advancing BCI-based rehabilitation strategies for unilateral upper limb recovery.