Gaussian wavelet based dynamic filtering (GWDF) method for medical ultrasound systems

Area of Science:

• Medical Imaging
• Signal Processing
• Biomedical Engineering

Background:

• Ultrasound echo signals are susceptible to noise, which can impede diagnostic accuracy.
• Traditional filtering methods may not adapt well to varying signal characteristics at different depths.
• Improving signal-to-noise ratio (SNR) is crucial for enhanced ultrasound image quality.

Purpose of the Study:

• To propose and evaluate a novel dynamic filtering method for noise reduction in ultrasound echo signals.
• To adapt filtering parameters in real-time based on signal characteristics at varying depths.
• To enhance the extraction of useful diagnostic information from ultrasound data.

Main Methods:

• A dynamic filtering approach utilizing Gaussian wavelet filters is developed.
• A mother wavelet is selected matching the transmitted signal's central frequency (CF) and frequency bandwidth (FB).
• Autocorrelation estimates the actual received signal frequency, enabling dynamic wavelet dilation for depth-specific filtering.

Main Results:

• The proposed method successfully removes noise from ultrasound echo signals.
• Simulations and experiments demonstrate optimal signal-to-noise ratio (SNR) achievement.
• Useful information extraction along the depth direction is maintained irrespective of object type.

Conclusions:

• The novel dynamic Gaussian wavelet filtering method effectively reduces noise in ultrasound echo signals.
• This technique allows for adaptive filtering, improving SNR and diagnostic information extraction.
• The method shows promise for enhancing the quality and reliability of ultrasound imaging.