Preliminary Comparison of Multi-scale and Multi-model Direct Inversion Algorithms for 3T MR Elastography

PURPOSE To elucidate whether any differences are present in the stiffness map obtained with a multiscale direct inversion algorithm (MSDI) vs that with a multimodel direct inversion algorithm (MMDI), both qualitatively and quantitatively. MATERIALS AND METHODS The MR elastography (MRE) data of 37 consecutive patients who underwent liver MR elastography between September and October 2014 were retrospectively analyzed by using both MSDI and MMDI. Two radiologists qualitatively assessed the stiffness maps for the image quality in consensus, and the measured liver stiffness and measurable areas were quantitatively compared between MSDI and MMDI. RESULTS MMDI provided a stiffness map of better image quality, with comparable or slightly less artifacts. Measurable areas by MMDI (43.7 ± 17.8 cm2) was larger than that by MSDI (37.5 ± 14.7 cm2) (P < 0.05). Liver stiffness measured by MMDI (4.51 ± 2.32 kPa) was slightly (7%), but significantly less than that by MSDI (4.86 ± 2.44 kPa) (P < 0.05). CONCLUSION MMDI can provide stiffness map of better image quality, and slightly lower stiffness values as compared to MSDI at 3T MRE, which radiologists should be aware of.