Quantitative Shear Wave Elastography: A Phantom-Based Comparison of Two Ultrasound Systems

Area of Science:

• Medical Imaging and Ultrasound Technology
• Biomedical Engineering and Instrumentation
• Diagnostic Ultrasound and Elastography

Background:

• Clinical standardization of shear wave elastography (SWE) is limited by inter-vendor variability in measurement consistency.
• Accurate assessment of tissue mechanical properties using SWE requires robust validation against known standards.
• Phantom-based studies are crucial for evaluating ultrasound system performance and identifying sources of measurement error.

Purpose of the Study:

• To establish a robotically controlled, phantom-based framework for evaluating cross-platform measurement consistency in SWE.
• To quantify inter-vendor variability and diagnostic threshold requirements for SWE systems.
• To interpret measurement bias in relation to physical phantom properties and imaging system characteristics.

Main Methods:

• A custom polyvinyl chloride-graphite phantom with eight spherical inclusions (15-25 mm diameter, 25-95 mm depth, 23.53-259.58 kPa stiffness) was used.
• Robotic automation ensured standardized probe positioning and contact for Samsung HS50 and Aixplorer ultrasound systems.
• Cross-platform reproducibility, accuracy against ground truth, and diagnostic threshold performance were assessed over 80 measurements.

Main Results:

• Both systems exhibited excellent intra-machine reproducibility (CV: Samsung 0.42%, Aixplorer 0.46%) and strong inter-machine correlation (r=0.9951).
• Substantial cross-platform differences were observed, with systematic bias (7.05 kPa) and wide limits of agreement (38.7 kPa).
• All inclusions were underestimated, with measurement distortion linked to inclusion size, depth, and stiffness; dynamic range compression was noted (Samsung: 68.7%, Aixplorer: 59.1%).

Conclusions:

• Despite high reproducibility, significant system-dependent bias prevents direct transferability of SWE measurements across ultrasound platforms.
• A validated methodology linking SWE bias to physical lesion properties and system characteristics is essential for reliable interpretation.
• System-specific calibration and collaboration are required to ensure accurate and consistent clinical application of SWE.