Related Experiment Video
Updated: Jun 23, 2026
07:47
Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
Published on: February 14, 2018
Automatic adaptive onset detection using an electromyogram with individual difference for control of a meal
1Department of Automation, East China University of Science and Technology, Shanghai, China. shuyee_zhang@yahoo.com.cn
Journal of Medical Engineering & Technology
|April 23, 2009
Summary
This study introduces a new method for detecting muscle activation using electromyogram (EMG) power with an adaptive threshold. This approach enables a more reliable EMG-controlled meal assistance robot for individuals with limb deficiencies.
Area of Science:
- Biomedical Engineering
- Rehabilitation Technology
- Human-Machine Interfaces
Background:
- Accurate detection of muscle activation onset from electromyogram (EMG) signals is crucial for effective prosthetic and assistive device control.
- Existing methods often struggle with individual variability in muscle contraction and resting states, leading to control inaccuracies.
- Developing robust EMG control systems is essential for improving the independence and quality of life for individuals with limb deficiencies.
Purpose of the Study:
- To propose and evaluate a novel EMG onset detection approach using EMG power with an automatic adaptive threshold.
- To develop an effective EMG-controlled meal assistance robot tailored for users with limb deficiencies.
- To address individual differences in muscle signal characteristics for improved control reliability.
Main Methods:
- A new method utilizing EMG power with an automatic adaptive threshold for muscle activation onset detection was developed.
- The adaptive threshold is dynamically set based on the latest EMG signal to account for individual variations in contraction and resting power.
- The proposed method was implemented and tested in the context of an EMG-controlled meal assistance robot.
Main Results:
- The proposed method demonstrated automatic adjustment capabilities, effectively avoiding false alarms during EMG signal processing.
- The system performed well even when variations in muscle contraction power were present.
- The adaptive threshold approach successfully compensated for individual differences in EMG signal characteristics.
Conclusions:
- The novel EMG onset detection method with an automatic adaptive threshold provides a robust solution for reliable muscle activation detection.
- The developed EMG-controlled meal assistance robot shows promise as an effective and comfortable human-machine interface for limb-deficient patients.
- This technology has the potential to significantly enhance assistive capabilities for individuals requiring support with daily tasks.
