Analysis of liver viscosity behavior as a function of multifrequency magnetic resonance elastography (MMRE) postprocessing

• Published in: Journal of magnetic resonance imaging Vol. 38; no. 2; pp. 422 - 428
• Main Authors: Leclerc, Gwladys E., Charleux, Fabrice, Robert, Ludovic, Ho Ba Tho, Marie-Christine, Rhein, Colette, Latrive, Jean-Paul, Bensamoun, Sabine F.
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01.08.2013; Wiley Subscription Services, Inc; Wiley-Blackwell
• Subjects: Adult; Algorithms; Bioengineering; Biomechanics; Computer Simulation; elasticity; Elasticity Imaging Techniques - methods; Engineering Sciences; Female; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Imaging; Life Sciences; Liver; Liver - pathology; Liver - physiopathology; Liver Cirrhosis - pathology; Liver Cirrhosis - physiopathology; liver fibrosis; Magnetic resonance imaging; Male; Mechanics; Middle Aged; Models, Biological; multifrequency magnetic resonance elastography; Reproducibility of Results; rheological models; Sensitivity and Specificity; Viscosity; Young Adult; liver fibrosis; viscosity; multifrequency magnetic resonance elastography; elasticity; rheological models
• ISSN: 1053-1807; 1522-2586; 1522-2586
• DOI: 10.1002/jmri.23986

Purpose To analyze the relevance of the viscosity measurement as a liver diagnostic marker. Materials and Methods To determine the level of fibrosis, a Fibroscan test was performed on 40 subjects (10 healthy volunteers and 30 patients). Subsequently, multifrequency magnetic resonance elastography (MMRE) tests were made with a pneumatic driver at 60, 70, and 80 Hz. Phase images were analyzed with two different postprocessing methods, without (Method 1) and with (Method 2) the inversion algorithm (IA), using rheological models (Voigt, springpot) in order to characterize the viscoelastic properties (viscosity: η and elasticity: μ). Results MRE cartography of the viscous tendency (G″MRE_M2) measured within the region of interest (ROI) of the liver increased as a function of the level of fibrosis. Similar results were also obtained for the viscosity (ηmodels_M1) calculated with a postprocessing without IA. However, the viscosity (ηmodels_M2) remained constant with the stage of fibrosis when the postprocessing was composed of an IA. The experimental (μMRE_M1 and G′MRE_M2) and rheological (μmodels_M2 and μmodels_M1) elasticities always increased with the level of fibrosis regardless of the postprocessing method. Conclusion The variation of the liver viscosity parameter as a function of postprocessing revealed that this parameter should be further investigated to demonstrate its relevance in clinical practice. J. Magn. Reson. Imaging 2013;38:422–428. © 2012 Wiley Periodicals, Inc.