Related Experiment Video
Updated: Jul 10, 2026
09:32
Cortical Source Analysis of High-Density EEG Recordings in Children
Published on: June 30, 2014
Two channel EEG thought pattern classifier
D A Craig1, H T Nguyen, H A Burchey
1Key University Research Centre for Health Technologies, Faculty of Engineering, University of Technology, Sydney, NSW, Australia.
Summary
This study introduces a real-time electro-encephalogram (EEG) system for hands-free wheelchair control. The system accurately interprets mental commands using a neural network, enabling intuitive device operation.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Human-Computer Interaction
Background:
- Traditional assistive technologies often require physical interaction, posing challenges for individuals with severe motor impairments.
- Developing intuitive and non-invasive control methods for assistive devices is crucial for enhancing user independence.
Purpose of the Study:
- To develop and evaluate a real-time electro-encephalogram (EEG) identification system for hands-free control of a powered wheelchair.
- To assess the system's accuracy and response time in classifying user mental commands.
Main Methods:
- Utilized a ProComp+ encoder to amplify and digitize EEG signals from two scalp electrodes (O(1) and C(4)).
- Implemented a real-time multilayer neural network for classifying EEG patterns corresponding to mental commands.
- Transferred digitized EEG data to a host computer via an RS232 interface for processing.
Main Results:
- The system demonstrated a rapid response, detecting changes in user thought patterns within 1 second.
- Achieved a classification accuracy exceeding 79% for three distinct mental commands (forward, left, right).
- Successfully enabled control of a powered wheelchair using only two EEG electrodes.
Conclusions:
- Real-time EEG analysis with neural networks offers a viable solution for hands-free control of assistive devices.
- The developed system shows significant potential for improving mobility and independence for individuals with disabilities.
- Further research can explore expanding the command set and improving accuracy for more complex interactions.
