Reconstruction accuracy dependence with induced-shear-wave magnitude in Magnetic Resonance Elastography

Since 1996, Magnetic Resonance Elastography (MRE) holds the promise for absolute quantitation of the mechanical parameters of living tissues [1]. The reproducibility of the technique was challenged [2] while the measurement precision was determined by the uncertainty of the recorded MR-signal phase onto which the inferred motion is encoded [3]. We assumed that the ratio of the resulting total wave amplitude to its related uncertainty, AT/ΔAT,_ _should be considered to validate the acquired set of MRE data. Nevertheless, as long as this ratio is greater than unity, the validity of the extracted mechanical parameters might not be questioned. Here, we extract the complex shear modulus, G=G′+G′′, by inversion of the three-dimensional equation of motion [4] for a wide range of inferred wave amplitude, starting from zero, in a breast phantom. The shear dynamic, G′, and loss, G′′, moduli were found to increase with the wave amplitude before reaching a plateau at ratios AT/ΔAT much greater than one. Experiments were carried with standard motion-sensitized refocused field echo (RFE) [1] and motion fractional-encoding fast field echo (FFE) [5], for which sensitivities largely differ, so the relevance of a MRE-validity threshold based on the ratio AT/ΔAT could be exhibited.