Comparison of shear velocity dispersion in viscoelastic phantoms measured by ultrasound-based shear wave elastography and magnetic resonance elastography

Abstract

Pathological processes in soft tissues cause changes in mechanical properties. Elastography methods have emerged to make quantitative measurements of the shear modulus or shear wave velocity (SWV) as a noninvasive way to provide diagnostic information. In an effort to standardize ultrasound-based measurements of SWV, the Radiological Society of North America Quantitative Imaging Biomarkers Alliance (RSNA QIBA) has established working groups to develop profiles for different biomarkers including the use of SWS for staging of patients with liver fibrosis. To understand how ultrasound-based measurements vary with tissue viscoelasticity, measurements were made in viscoelastic phantoms with ultrasound-based shear wave elastography (US-SWE) and magnetic resonance elastography (MRE) over a wide frequency range. Three different viscoelastic phantoms were tested with US-SWE and MRE and analysis was considered over a range of frequencies (60–600 Hz). MRE was performed with an electro-mechanical driver driven with continuous vibration at various frequencies. Shear wave motion was measured with a phase-sensitive pulse sequence and the complex modulus and SWS at each frequency was estimated. The SWV in each phantom were fit with a power-law. US-SWE was performed with a Verasonics system and curved array transducer. An acoustic radiation force push of 800 μs was focused at 30, 45, and 70 mm and shear wave motion was measured and two-dimensional Fourier analysis was used to measure SWS dispersion. The agreement was generally good and the percent differences for a focal depth of 45 mm ranged from − 6.35–11.99% over the range of 60–300 Hz. The results for a focal depth of 30 mm had percent differences ranging from −4.37–12.17% from 60–300 Hz, and for a focal depth of 70 mm −1.05–16.46% from 80–200 Hz. The percent differences were typically above 15% at frequencies of 400–600 Hz. These measurements showed very good agreement between US-SWE and MRE methods in viscoelastic phantoms as MRE is being considered as a QIBA reference standard for US-SWE methods for liver SWE measurements.