Minimizing Measurement-Induced Errors in Viscoelastic MR Elastography.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Mechanics

Background:

• Magnetic Resonance Elastography (MRE) is crucial for tissue elasticity imaging.
• Measurement noise significantly impacts MRE results, affecting diagnostic accuracy.
• Existing MRE methods are sensitive to noise, particularly its indirect effects via boundary conditions.

Purpose of the Study:

• To develop a novel MRE formulation that mitigates noise influence for improved accuracy.
• To address the sensitivity of the inverse problem in MRE to measurement data.
• To enhance the reliability of tissue elasticity mapping.

Main Methods:

• Application of a coupled adjoint field formulation for viscoelastic parameter identification.
• Avoidance of indirect noise influence through applied boundary conditions.
• Development of a well-posed coupled field problem using adjoint field conditions.
• Implementation of a nearly incompressible, parallel subdomain-decomposition approach.
• Numerical simulation studies for method verification and performance exploration.

Main Results:

• The novel formulation effectively avoids indirect noise influence from boundary conditions.
• The adjoint field conditions relieve computed displacement fields from kinematic boundary errors.
• Numerical simulations demonstrate the method's robustness and performance.
• Successful in-vivo demonstration in the human brain, yielding viable tissue property maps.

Conclusions:

• The coupled adjoint field formulation offers a robust approach to MRE.
• This method enhances accuracy and provides viable tissue property maps, even in challenging configurations.
• The findings advance the field of quantitative tissue elasticity imaging.