Related Experiment Video
Updated: Jul 1, 2025
06:58
A Structured Rehabilitation Protocol for Improved Multifunctional Prosthetic Control: A Case Study
Published on: November 6, 2015
9.5K
A novel brain-controlled prosthetic hand method integrating AR-SSVEP augmentation, asynchronous control, and machine
Xiaodong Zhang1,2, Teng Zhang3,2, Yongyu Jiang1
1School of Mechanical Engineering, Xi'an Jiaotong University, Xi'an, Shannxi, 710049, China.
Heliyon
|March 11, 2024
Summary
This study introduces an augmented reality (AR) brain-computer interface (BCI) for prosthetic hands, enhancing control with a novel stimulus paradigm and pattern recognition algorithm. The system demonstrated improved accuracy and practical usability for disabled patients.
Area of Science:
- Biomedical Engineering
- Neuroscience
- Human-Computer Interaction
Background:
- Brain-computer interface (BCI) systems using steady-state visual evoked potentials (SSVEP) offer potential for prosthetic hand assistance in disabled individuals.
- Existing BCI systems face limitations in interaction, stimulus paradigms, and control logic, hindering practical application.
Purpose of the Study:
- To innovate the visual stimulus paradigm and asynchronous decoding/control strategy for BCI-controlled prosthetic hands by integrating augmented reality (AR) technology.
- To develop an advanced asynchronous pattern recognition algorithm to improve the interaction logic and practical capabilities of BCI systems.
Main Methods:
- An AR-based asynchronous visual stimulus paradigm with 8 control modes was developed.
- A novel asynchronous pattern recognition algorithm, Center-ECCA-SVM, combining canonical correlation analysis and support vector machines, was proposed.
- An intelligent BCI system switch utilizing the YOLOv4 deep learning object detection algorithm was implemented for enhanced user interaction.
Main Results:
- The AR paradigm increased SSVEP spectrum amplitude by 17.41% and signal-noise ratio (SNR) by 3.52% compared to LCD.
- The Center-ECCA-SVM classifier achieved high asynchronous pattern recognition accuracy (94.66%–97.40%) within a 2s analysis time.
- The YOLOv4-tiny model enabled real-time prosthetic hand detection at 25.29fps with 96.4% confidence, facilitating completion of daily tasks.
Conclusions:
- The developed AR-BCI system significantly improves user interaction for prosthetic hand control.
- The study validates the effectiveness and practicality of the proposed system for alternative prosthetic control and rehabilitation programs.
