Multidimensional Clutter Filter Optimization for Ultrasonic Perfusion Imaging.

Area of Science:

• Ultrasound Imaging
• Biomedical Engineering
• Medical Physics

Background:

• Conventional power Doppler (PD) imaging often requires exogenous contrast agents to visualize peripheral blood perfusion.
• Existing methods face challenges in distinguishing slow blood flow from clutter signals, limiting sensitivity and specificity.
• Improving the detection of subtle perfusion patterns is crucial for accurate diagnosis and monitoring of various conditions.

Purpose of the Study:

• To develop and validate novel pulse-echo acquisition and clutter filtering techniques for enhanced PD imaging.
• To increase the sensitivity and specificity of PD imaging for detecting slow, disorganized peripheral blood flow.
• To reduce or eliminate the need for exogenous contrast agents in peripheral perfusion assessment.

Main Methods:

• Acquisition of echoes using sparse Doppler pulse transmissions over extended durations to capture slow motion.
• Application of 3-D singular-value decomposition (SVD) clutter filtering by arranging temporal and spatial echo data.
• Development of a statistical classifier for partitioning the 3-D SVD core tensor to separate blood and clutter subspaces.

Main Results:

• The proposed technique successfully separates blood echoes from clutter, even when subspaces do not fully uncouple.
• Validated using simulations, flow phantoms, and in vivo murine melanoma models.
• Routine mapping of phantom flows and tumor perfusion signals at speeds below 3 mL/min achieved with narrow eigen-bandwidth clutter.

Conclusions:

• The developed method significantly enhances PD imaging for peripheral perfusion assessment.
• It offers improved sensitivity and specificity, enabling visualization of low-flow states without contrast agents.
• The technique is well-suited for clinical applications requiring detailed peripheral perfusion imaging.