Variational mode decomposition based ECG denoising using non-local means and wavelet domain filtering

Area of Science:

• Biomedical Engineering
• Signal Processing

Background:

• Existing electrocardiogram (ECG) denoising methods struggle to eliminate noise across the full frequency spectrum.
• Effective noise reduction is crucial for accurate ECG signal interpretation and diagnosis.

Purpose of the Study:

• To develop a novel and effective ECG denoising approach.
• To overcome the limitations of traditional denoising techniques like Discrete Wavelet Transform (DWT) and Non-Local Means (NLM) estimation.

Main Methods:

• The proposed method utilizes Variational Mode Decomposition (VMD) to decompose ECG signals into narrow-band variational mode functions (VMFs).
• Higher frequency VMFs are filtered using DWT-thresholding, while lower frequency VMFs are denoised with NLM estimation.
• The non-recursive nature of VMD allows for parallel processing of NLM and DWT, enhancing efficiency.

Main Results:

• The combined NLM and DWT approach effectively addresses the individual limitations of each technique.
• Signal reconstruction using denoised VMFs results in a cleaner ECG signal.
• Simulations on the MIT-BIH Arrhythmia database demonstrate superior performance compared to existing state-of-the-art methods.

Conclusions:

• The proposed VMD-based ECG denoising method offers a significant improvement over current techniques.
• This approach provides a robust solution for accurate ECG signal acquisition and analysis.
• The synergistic application of VMD, NLM, and DWT enhances denoising performance and diagnostic utility.