K Fujimoto*, T Shiinoki, Y Yuasa, Y Kawazoe, M Yamane, T Sera, H Tanaka, Yamaguchi University

• 
  K Fujimoto

• 
  

TH-C-TRACK 3-3 (Thursday, 7/29/2021) 1:00 PM - 2:00 PM [Eastern Time (GMT-4)]

Purpose: Finite element method (FEM) is a promising method to calculate the biomechanical parameters of human liver by performing the stress analysis. However, its performance to evaluate the distribution of liver fibrosis has not been demonstrated. In this study, to assess the performance to evaluate the distribution of liver fibrosis, the liver elasticity map estimated by using FEM (Elasticity-map) and the MRE were compared.

Methods: Twelve patients who underwent magnetic resonance imaging (MRI) and MRE were enrolled in this study. The axial T1-weighted MR images were acquired at shallow inspiration and shallow expiration breath-hold. The displacement vector field (DVF) between T1-weighted MR images at the inspiration and expiration were calculated using deformable image registration. Elasticity-maps were calculated by using FEM and DVF. MRE images were acquired at shallow expiration breath-hold. The grade of liver fibrosis in all patients were determined by radiologist based on liver elasticity measured by MRE. The whole and regional liver (right lobe and left lobe of liver) elasticity estimated via Elasticity-map were evaluated for correlations with the elasticity obtained via MRE.

Results: The whole liver elasticity estimated via Elasticity-map was significantly correlated with the elasticity measured via MRE (r = 0.950, p<0.001) and the liver fibrosis grade (r = 0.915, p< 0.001). Furthermore, for the regional liver elasticity, the correlation coefficient between the Elasticity-map and MRE was 0.910 (p<0.001) for right lobe of liver and 0.905 (p<0.001) for left lobe of liver.

Conclusion: The elasticity for whole liver, right lobe and left lobe of liver estimated via Elasticity-map were strongly correlated with the elasticity measured via MRE. Therefore, the Elasticity-map has a potential to be a feasible imaging biomarker to evaluate the distribution of liver fibrosis using only T1-weighted images without additional inspection and costs for special equipment for elastography.

Funding Support, Disclosures, and Conflict of Interest: Japan society for the promotion of science (JSPS) KAKENHI Grant number 20K16789(KF), 18K07753(TS)

Handouts

Keywords

Finite Element Analysis, MR Elastography, Image Correlation

Taxonomy

IM/TH- Mathematical/Statistical Foundational Skills: Finite elements analysis

Contact Email