Validation of Motion-Robust Liver Diffusion MRI in Multi-Stiffness Pulsatile Motion Phantoms

Motivation: Quantitative diffusion MRI is a proposed marker for assessment of liver fibrosis. However, poor reproducibility and lack of highly controlled validation of liver ADC mapping precludes its clinical utilization. Goal(s): Introduce hydrogel liver models with pulsatile motion and varying stiffness. These enable controlled validation of ADC accuracy and reproducibility across DWI acquisition parameters and physiological-mimicking motion. Approach: Conventional monopolar (MONO) and motion-robust M1-optimized diffusion waveforms (MODI) were used to acquire DWI of three hydrogel liver models. Results: MODI-DWI resulted in less biased DWI and ADC maps than MONO-DWI in areas of motion. A significant inverse relationship was observed between ADC and phantom stiffness. Impact: Quantitative diffusion MRI may enable assessment of liver fibrosis. However, the relationship between diffusion parameters and stiffness requires controlled evaluation. The proposed phantom-based approach may help validate and optimize diffusion MRI of the liver and other abdominal organs.