Circular statistics vector for improving coherent plane wave compounding image in Fourier domain

Area of Science:

• Ultrasound imaging
• Medical imaging signal processing
• Beamforming algorithms

Background:

• High frame rate ultrasound imaging is crucial for dynamic processes.
• Conventional beamforming methods face trade-offs between image quality and computational complexity.
• Existing speckle reduction techniques can sometimes degrade image resolution.

Purpose of the Study:

• To introduce a Fourier domain circular statistics vector (FD-CSV) weighting method for ultrasound plane-wave imaging.
• To evaluate the performance of FD-CSV in terms of image quality (resolution, contrast) and computational load.
• To compare FD-CSV with conventional methods like DAS-CSV and variance-based speckle reduction techniques (CF, PCF).

Main Methods:

• Extraction of cosine and sine components from undelayed RF signals.
• Application of Fourier domain beamforming to these components.
• Construction of the circular statistics vector (CSV) using FD beamformed components.
• Weighting the FD beamformed amplitude image with the derived CSV.

Main Results:

• The proposed FD-CSV method achieved significant reduction in computational load compared to DAS-CSV.
• FD-CSV demonstrated comparable or improved image quality (resolution, contrast) to conventional methods.
• CSV based on mean resultant vector calculation showed better speckle preservation and tolerance to phase errors than variance-based methods.

Conclusions:

• The FD-CSV weighting method offers a promising approach for high-quality, high frame rate ultrasound imaging.
• This method effectively balances image quality enhancement with reduced computational complexity.
• FD-CSV provides a robust alternative to existing beamforming and speckle reduction techniques in ultrasound.