Related Experiment Video
Updated: Aug 6, 2026
06:11
Brain-Computer Interface-controlled Upper Limb Robotic System for Enhancing Daily Activities in Stroke Patients
Published on: April 18, 2025
An Online Bimanual EEG-MI-BCI With Shared Control for Bilateral Robotic-Assisted Training
Summary
This study introduces a new brain-computer interface (BCI) for robotic rehabilitation, enabling simultaneous control of both hands through motor imagery (MI). The system shows promise for improving rehabilitation outcomes by accurately decoding intended movements.
Area of Science:
- Neuroscience and Biomedical Engineering
- Rehabilitation Technology
- Brain-Computer Interfaces
Background:
- Bimanual motor tasks are crucial for daily activities and rehabilitation.
- Existing brain-computer interfaces (BCIs) for rehabilitation primarily focus on single-limb control.
- Accurate and robust decoding of electroencephalography (EEG) signals remains a challenge for BCI-driven robotic systems.
Purpose of the Study:
- To develop and evaluate a novel EEG-based BCI system for online bilateral robot-assisted training.
- To investigate a bimanual motor imagery (MI) paradigm involving coordinated movements of both hands.
- To implement a shared control strategy for error correction and autonomous robot movement management.
Main Methods:
- Developed a bimanual EEG-MI paradigm with three coordinated movement directions (left, middle, right) for both hands.
- Employed a shared control strategy combining prior knowledge-based assistance and robot autonomy.
- Conducted offline and online decoding experiments using six common models (EEGNet being optimal) and a multi-step reaching task.
Main Results:
- Offline decoding accuracy significantly surpassed the chance level (33.33%), with EEGNet achieving 52.93%.
- Online decoding with EEGNet reached an accuracy of 49.67%.
- Assistance levels (low, moderate, high) improved task success rates from 48.33% to 71.67%, 80.00%, and 90.00%, respectively.
Conclusions:
- The developed EEG-MI-BCI system demonstrates online feasibility for decoding coordinated bimanual movements.
- The shared control strategy effectively enhances task performance in robot-assisted training.
- This system holds potential for advancing bilateral robot-assisted rehabilitation therapies.
