Related Experiment Video
Updated: Jun 26, 2026
09:42
Acquisition and Semi-Automated Analysis of Respiratory Muscle Surface Electromyography
Published on: January 24, 2025
Compression of electromyographic signals using image compression techniques
Marcus Vinícius Chaffim Costa1, Pedro de Azevedo Berger, Adson Ferreira da Rocha
1Department of Electrical Engineering, University of Brasília, DF, Brazil. chaffim@gmail.com
Summary
This study introduces a novel algorithm for compressing electromyographic (EMG) signals using the JPEG2000 system. The method achieves high compression ratios for both isometric and isotonic contractions, proving effective for long-term EMG data storage and transmission.
Area of Science:
- Biomedical Engineering
- Signal Processing
- Data Compression
Background:
- Electromyographic (EMG) signal transmission and storage are crucial for long-term monitoring.
- Limited research exists on effective compression techniques for EMG data.
- Efficient compression is needed to manage large volumes of physiological signal data.
Purpose of the Study:
- To develop and evaluate a compression algorithm for electromyographic (EMG) signals.
- To adapt the JPEG2000 coding system, originally for images, for EMG signal compression.
- To assess the algorithm's performance for both isometric and isotonic muscle contractions.
Main Methods:
- An algorithm based on the JPEG2000 coding system was developed for EMG signal compression.
- The algorithm was tested on EMG signals acquired during isometric and isotonic contractions.
- Performance was evaluated using compression factors and Percentage Residual Difference (PRD).
- Results were compared against other wavelet transform-based compression algorithms.
Main Results:
- The JPEG2000-based algorithm achieved compression factors of 75-90% for both isometric and isotonic EMG signals.
- For isometric contractions, average PRD ranged from 3.75% to 13.7%.
- For isotonic contractions, average PRD ranged from 3.4% to 7%.
Conclusions:
- The JPEG2000 algorithm is effective for compressing electromyographic signals.
- This method offers significant data reduction for long-term EMG signal storage and transmission.
- The algorithm demonstrates high compression efficiency with acceptable signal fidelity for various contraction types.
