Simultaneous powerline interference and baseline wander removal from ECG and EMG signals by sinusoidal modeling

Area of Science:

• Biomedical Engineering
• Signal Processing

Background:

• Biopotential signals are often corrupted by powerline interference (PLI) and baseline wander (BW).
• Accurate denoising is crucial for reliable signal interpretation in medical diagnostics.

Purpose of the Study:

• To develop a compact and efficient joint modeling approach for PLI and BW.
• To improve the denoising of biopotential signals, specifically electrocardiographic (ECG) and electromyographic (EMG) signals.

Main Methods:

• A novel model representing PLI and BW using harmonically related sinusoids modulated by time polynomials.
• Linear-in-parameter model solved efficiently via linear least-squares estimation.
• Evaluation on real ECG and EMG data against established denoising techniques.

Main Results:

• The proposed joint modeling method effectively denoises biopotential signals.
• Demonstrated superior performance compared to three reference methods in reducing residual interference energy.
• Successfully captured harmonicity, frequency, and amplitude deviations of interference and wander.

Conclusions:

• The presented compact approach offers a robust solution for simultaneous PLI and BW removal.
• This method enhances the quality of denoised biopotential signals for clinical applications.
• The linear least-squares solution provides an efficient and practical denoising tool.