Competitive Swarm Optimized SVD Clutter Filtering for Ultrafast Power Doppler Imaging

Area of Science:

• Medical imaging
• Ultrasound technology
• Signal processing

Background:

• Ultrafast power Doppler imaging (uPDI) enhances sensitivity for small vascular structures.
• Clutter filtering, often using Singular Value Decomposition (SVD), is crucial for uPDI.
• Existing SVD filters with two cutoffs struggle to optimally separate tissue, blood, and noise.

Purpose of the Study:

• To introduce a novel Competitive Swarm-Optimized SVD (CSO-SVD) clutter filter.
• To improve the quality and diagnostic potential of uPDI.
• To overcome limitations of traditional SVD-based clutter filtering methods.

Main Methods:

• Developed a CSO-SVD clutter filter that avoids dual cutoffs.
• Employed Competitive Swarm Optimization (CSO) to identify blood signal components within singular values.
• Validated the CSO-SVD filter using public in vivo ultrasound datasets.

Main Results:

• The CSO-SVD filter demonstrated superior performance compared to state-of-the-art SVD filters.
• Achieved significant improvements in Contrast-to-Noise Ratio (CNR), Signal-to-Noise Ratio (SNR), and Blood-to-Clutter Ratio (BCR).
• Specifically, in rat brain bolus imaging, improvements were observed: CNR (+0.99 ± 0.08 dB), SNR (+0.79 ± 0.08 dB), and BCR (+1.95 ± 0.03 dB).

Conclusions:

• The proposed CSO-SVD clutter filter effectively suppresses clutter and noise while preserving vital blood signals in uPDI.
• This advanced filtering technique offers a better balance for enhanced vascular visualization.
• The CSO-SVD method represents a significant advancement in ultrasound clutter filtering for improved diagnostic accuracy.