Measurement consistency of MR elastography at 3.0 T: comparison among three different region-of-interest placement methods.

Purpose: To test inter- and intraobserver consistency of liver stiffness measurement on MR elastography (MRE) at 3.0 T.

Materials And Methods: Two abdominal radiologists independently measured stiffness of the liver on MRE in three volunteers and seven patients with chronic liver diseases using three different region-of-interest (ROI) placement methods. Methods 1 and 2 involved placing circular and free-hand-drawn ROIs, respectively, visually referring to anatomical (three-dimensional T1-weighted) and wave images. Method 3 involved placing ROIs on the fused images of MRE and anatomical images developed on a work station, visually referring to wave images. The inter- and intraobserver consistency was assessed with regression and Bland-Altman analysis.

Results: Thirty-eight images were available for measurement in total. As for interobserver consistency, method 3 showed the best regression coefficient, correlation coefficient, and y intercept. The absolute values of the interobserver differences for method 3 were significantly smaller than those of method 1 or method 2 (p < 0.05, each). Intraobserver consistency of method 3 was excellent for both observers.

Conclusion: Stiffness measurement of the liver on MRE performed with the fusion method at 3.0 T provides the highest inter- and intraobserver consistency.