High-pass filtering to remove electrocardiographic interference from torso EMG recordings.

Area of Science:

• Biomedical Engineering
• Neuroscience
• Biomechanics

Background:

• Electromyographic (EMG) signals from torso muscles are crucial for biomechanical models.
• Electrocardiographic (ECG) interference frequently contaminates these EMG recordings.
• High-pass filtering is a common but under-investigated method to reduce ECG contamination.

Purpose of the Study:

• To investigate the impact of high-pass filter cut-off frequencies on EMG signals from torso muscles.
• To determine the optimal cut-off frequency for reducing ECG contamination while preserving EMG signal quality.

Main Methods:

• Surface EMGs were recorded from rectus abdominis, external oblique, and erector spinae muscles in five subjects.
• Signals were digitally high-pass filtered at 10, 30, and 60 Hz.
• Integrated EMG and power analyses were performed on the filtered signals.

Main Results:

• Increasing the cut-off frequency reduced ECG contamination and smoothed the EMG signal.
• Higher cut-off frequencies also decreased the EMG signal amplitude.
• A cut-off frequency of approximately 30 Hz was found to be optimal.

Conclusions:

• High-pass filtering is an effective technique for reducing ECG interference in integrated EMG recordings.
• The choice of cut-off frequency is critical; inappropriate settings can lead to incomplete artifact removal or excessive signal filtering.
• Approximately 30 Hz is recommended as an optimal cut-off frequency for torso muscle EMG analysis.