A single-ensemble-based hybrid approach to clutter rejection combining bilinear Hankel with regression

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Signal Processing

Background:

• Ultrasound Doppler echoes from soft tissue (clutter) impede accurate blood flow measurement in Color Flow Imaging (CFI).
• Existing multi-ensemble clutter filters can reduce image resolution and frame rate.
• Single-ensemble filters often lack the accuracy needed for both low- and high-velocity blood flow estimation.

Purpose of the Study:

• To develop an advanced clutter filter for ultrasound Color Flow Imaging (CFI).
• To enhance the accuracy of blood flow estimation by effectively rejecting soft tissue clutter.
• To overcome limitations of existing single-ensemble clutter rejection filters.

Main Methods:

• A hybrid approach combining a low-order regression filter with Bilinear Hankel-SVD (B-Hankel-SVD) for signal decomposition.
• Utilizing B-Hankel-SVD for more precise decomposition of ultrasound Doppler signals compared to conventional Hankel-SVD.
• Extracting blood flow components based on a frequency threshold after signal decomposition.

Main Results:

• The proposed B-Hankel-SVD filter demonstrated reduced blood flow estimation bias compared to conventional Hankel-SVD.
• The hybrid algorithm proved more effective than standalone regression or Hankel-SVD filters in clutter rejection using single-ensemble samples.
• Significant improvements in blood flow frequency estimation and reduced estimation variance were observed with the hybrid method.

Conclusions:

• The hybrid B-Hankel-SVD filter offers superior performance in clutter rejection for ultrasound Doppler signals.
• This method enhances the accuracy of blood flow velocity and frequency estimation in CFI.
• The developed filter addresses key limitations of current clutter filtering techniques in medical ultrasound.