A hybrid denoising approach for PPG signals utilizing variational mode decomposition and improved wavelet...

Area of Science:

• Biomedical Engineering
• Signal Processing
• Physiological Monitoring

Background:

• Photoplethysmography (PPG) signals are susceptible to motion artifacts (MA) and baseline drift, compromising measurement accuracy.
• These interferences significantly degrade the reliability of PPG-based health monitoring.

Purpose of the Study:

• To develop an effective denoising algorithm for PPG signals.
• To eliminate baseline drift and high-frequency noise while preserving critical signal components (1-10 Hz).

Main Methods:

• A hybrid denoising approach combining Variational Mode Decomposition (VMD) and an improved wavelet threshold function.
• VMD decomposes PPG signals into intrinsic mode functions (IMFs) to remove baseline drift.
• An enhanced wavelet thresholding algorithm is applied to eliminate high-frequency noise.

Main Results:

• The hybrid method significantly improved signal-to-noise ratio (SNR) by 11.47% and reduced root mean square error (RMSE) by 26.75% compared to soft thresholding.
• Further improvements in SNR (15.54%) and RMSE (37.43%) were achieved compared to using the improved threshold function alone.
• Validation was performed on real-world PPG data, simulated signals, and the MIMIC-III database.

Conclusions:

• A robust PPG denoising algorithm effectively removes low-frequency baseline drift and high-frequency noise.
• The method preserves essential morphological characteristics of PPG signals.
• This technique provides a foundation for advanced time-domain feature extraction and model development in PPG analysis.