MR elastography inversion by compressive recovery.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Rheology

Background:

• Direct inversion (DI) for magnetic resonance elastography (MRE) faces challenges with data quality and noise, especially for low signal-to-noise ratio environments like lung MRE.
• Estimating tissue shear stiffness accurately is crucial for diagnosing conditions like liver fibrosis and assessing lung mechanics.

Purpose of the Study:

• To introduce and validate a novel MRE inversion technique, MRE inversion by compressive recovery (MICRo), designed for improved numerical stability and noise robustness.
• To compare the performance of MICRo against conventional DI methods in simulated phantoms, in vivo liver MRE, and in vivo lung MRE.

Main Methods:

• Development of MICRo, an inversion strategy leveraging prior knowledge of data noise and stiffness map sparsity.
• Validation using simulated phantoms with varying stiffness and noise levels.
• Comparative analysis with standard clinical multi-modal DI (MMDI) in healthy subjects and liver fibrosis patients, and in healthy subjects for lung MRE.

Main Results:

• MICRo demonstrated superior performance over filtered DI in simulated phantoms, especially under high stiffness and noise conditions.
• MICRo showed agreement with MMDI in hepatic MRE applications.
• In vivo lung MRE, MICRo provided stable stiffness estimations at different lung volumes, outperforming filtered DI.

Conclusions:

• MICRo offers a more stable and robust approach to MRE stiffness estimation compared to conventional DI, particularly in low SNR conditions.
• The technique shows significant potential for improving diagnostic accuracy in liver fibrosis and for in vivo lung stiffness assessment.
• Further clinical validation in larger populations is warranted to fully establish the clinical utility of MICRo.