Development of three-dimensional integral-type reconstruction formula for magnetic resonance elastography

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Rheology

Background:

• Tissue viscoelasticity, including storage and loss modulus, is linked to various diseases.
• Magnetic Resonance Elastography (MRE) is a non-invasive technique for quantifying tissue viscoelasticity.
• Conventional MRE inversion algorithms are susceptible to noise, degrading estimation accuracy.

Purpose of the Study:

• To develop novel inversion algorithms for MRE that are robust to noise in elastic wave images.
• To improve the accuracy of viscoelasticity estimation in MRE, particularly in the presence of background noise.

Main Methods:

• Proposed algorithms utilize the Voigt-type viscoelastic equation and Green's formula to enhance noise robustness.
• A stabilized curl-operator was implemented to mitigate compression waves in measurement data.
• Algorithm performance was evaluated using numerical simulations (isotropic and anisotropic) and phantom experiments.

Main Results:

• Numerical simulations showed normalized stiffness errors of 3% or less with the proposed algorithms.
• The novel algorithms demonstrated superior performance compared to conventional methods, especially with noisy data.
• Gel phantom experiments corroborated the simulation findings, confirming the algorithms' noise robustness.

Conclusions:

• Novel, quantitative, and noise-robust inversion algorithms for MRE have been successfully developed and validated.
• These algorithms accurately estimate storage and loss modulus, irrespective of noise, voxel anisotropy, and wave propagation direction.
• The developed algorithms are suitable for diverse three-dimensional MRE systems, advancing quantitative tissue characterization.