Unsupervised Beamforming with Optimized Coherence Loss for Clutter Suppression in Single Plane-Wave Ultrasound...

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Signal Processing

Background:

• Single plane-wave ultrasound imaging (SPWI) offers high acquisition speeds (>1000 Hz) for real-time applications.
• SPWI is limited by clutter artifacts like multipath reverberations, degrading image contrast and reliability.
• Existing methods struggle to effectively mitigate these artifacts while preserving temporal resolution.

Purpose of the Study:

• To develop an unsupervised beamforming method to suppress reverberation artifacts in SPWI.
• To enhance image contrast and diagnostic reliability in real-time ultrasound.
• To introduce an adaptive approach for optimizing coherence loss calculation.

Main Methods:

• Proposed an unsupervised beamforming approach with optimized deep coherence loss (UBF-DCLopt).
• Implemented adaptive signal coherence computation based on inter-frame decorrelation of plane-wave data.
• Utilized physics-based criteria to adaptively determine optimal plane-wave frames for coherence loss calculation, considering steering angle and pulse characteristics.
• Validated the method using simulations, phantom studies, and in vivo data.

Main Results:

• UBF-DCLopt demonstrated significant improvements in contrast-to-noise ratio (CNR).
• Achieved 22% CNR improvement in phantom experiments and 32% CNR improvement in in vivo studies compared to a fixed deep coherence loss method (UBF-DCL).
• Effectively suppressed reverberation artifacts while maintaining high spatiotemporal resolution.

Conclusions:

• The physics-informed unsupervised approach effectively suppresses reverberation artifacts in SPWI.
• The method enhances diagnostic accuracy in real-time ultrasound imaging.
• UBF-DCLopt offers a promising solution for improving the quality of high-speed ultrasound imaging.