The effects of geometry on stiffness measurements in high-field magnetic resonance elastography: A study on rodent cardiac phantoms

Cardiac magnetic resonance elastography (MRE) can be used to assess myocardial stiffness in vivo. Rodents play an important role in modern cardiovascular research, and small animal cardiac MRE may reveal important aspects of myocardial stiffness. The aim of this study was to explore the feasibility of small animal cardiac MRE through investigation of stiffness measurements of small cardiac phantoms that have known underlying stiffness. Agarose gel phantoms of three different geometrical designs were used: homogeneous gels, solid hearts, and biventricular phantoms. The size of the heart phantoms was comparable with that of an end-diastolic rat heart. All phantoms were made with different underlying stiffnesses agarose concentration, (7.5, 10.0,15.0)g/l, and MRE acquisition was performed with three different frequencies (360, 380, 400)Hz. Two different post-processing methods were applied to the MRE wave images: local frequency estimate (LFE) and direct inversion (DI). The stiffness associated with the different agarose concentrations (7.5, 10.0, 15.0)g/l in the homogenous gels at 400 Hz were (1.80 ± 0.18, 3.13 ± 0.20, 4.13 ± 0.37)kPa for LFE and (2.25 ± 0.24, 4.35 ± 0.45, 6.54 ± 0.44)kPa for DI, respectively. Significant differences in MRE-derived stiffness were observed among phantoms with different agarose concentrations for all geometries. However, biases in the stiffness measurements among the different geometries were observed and could not be explained by the measurement variability. The relative stiffness uncertainty was smallest for the LFE inversion algorithm. The stiffness measurements validate the use of the MRE technique to differentiate between various underlying stiffnesses in small cardiac phantoms. The stiffness measurements seemed to be dominated by geometrical effects when the cardiac MRE wavelength was longer than half the size of the heart. LFE was the inversion algorithm that was most sensitive to the changes in underlying stiffness.