Optimization of the Entropy-Based Wavelet Method for Removing Strong RF and AC Interferences in a Charge Detection...

Area of Science:

• Analytical Chemistry
• Spectrometry
• Signal Processing

Background:

• Linear ion trap mass spectrometry (LIT-MS) faces interference from radio frequency (RF) and alternating current (AC) fields.
• Charge sensing particle detectors (CSPD) are sensitive to these electromagnetic interferences, impacting data quality.
• Existing noise reduction methods may lack robustness or require arbitrary parameter selection.

Purpose of the Study:

• To develop and validate an entropy-based wavelet method for denoising mass spectra acquired using LIT-MS coupled with CSPD.
• To optimize wavelet parameters using energy-to-Shannon entropy for effective interference removal.
• To enhance the signal-to-noise ratio (S/N) for high-mass protein analysis.

Main Methods:

• An entropy-based wavelet denoising technique was developed, optimizing mother wavelet family and decomposition level via energy-to-Shannon entropy.
• Threshold values were determined using the median of sub-band coefficients at each decomposition level.
• Rigid threshold criteria were applied across all levels to eliminate noise and avoid arbitrary choices.

Main Results:

• The entropy wavelet method successfully denoised high-mass protein mass spectra.
• Significant S/N improvements were observed for immunoglobulin G (IgG) and alpha-2-macroglobulin (A2M) ions (68.03% and 81.73%, respectively).
• The method outperformed orthogonal wavelet packet decomposition (OWPD) filtering in noise reduction.

Conclusions:

• The developed entropy-based wavelet method provides effective interference removal in LIT-MS.
• This technique enhances the sensitivity and data quality for analyzing high-mass proteins.
• The optimized, non-arbitrary thresholding approach offers a robust solution for spectral denoising.