Electromyographic signal compression based on preprocessing techniques.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Data Compression

Background:

• Electromyographic (EMG) signals are often converted into 2D arrays for compression using image encoders.
• This conversion typically results in the loss of signal segment correlations, reducing compression efficiency.
• Existing methods struggle to maintain the integrity of temporal dependencies within EMG data during compression.

Purpose of the Study:

• To introduce novel preprocessing techniques for encoding EMG signals as 2D matrices.
• To enhance the exploitation of intersegment dependencies in EMG data.
• To improve the compression efficiency and quality of processed EMG signals.

Main Methods:

• Development of percentage difference sorting and relative complexity sorting for EMG signal preprocessing.
• Application of these techniques to real isometric EMG recordings.
• Compression of preprocessed signals using the JPEG2000 encoder.

Main Results:

• The proposed preprocessing methods effectively preserve intersegment dependencies in EMG signals.
• Experimental results demonstrate superior performance compared to existing state-of-the-art compression schemes.
• The framework achieves a favorable trade-off between compression ratio and distortion (PRD).

Conclusions:

• The novel sorting techniques offer a significant advancement in EMG signal compression.
• This approach effectively addresses the limitations of conventional 2D array conversion for EMG data.
• The proposed framework provides a more efficient and effective method for compressing biomedical signals.