Cross-Vendor Validation of Proton Density Fat Fraction and T₁ Mapping Using a Combined Proton Density Fat Fraction-T1...

Area of Science:

• Medical Imaging
• Biomarker Discovery
• Liver Disease Research

Background:

• Chronic liver disease involves coexisting fat and fibroinflammatory changes.
• Biomarkers proton density fat fraction (PDFF) and T1 are typically assessed separately.
• Reproducibility of these biomarkers under mutual confounding is not well understood.

Purpose of the Study:

• To evaluate the multicenter, multi-vendor reproducibility of chemical shift-encoded (CSE)-MRI-based PDFF mapping.
• To assess the reproducibility of MOLLI-based T1 mapping.
• To investigate reproducibility using a combined PDFF-T1 phantom accounting for confounding factors.

Main Methods:

• A prospective phantom study utilizing a commercial PDFF-T1 phantom with varying fat (0%-30%) and T1 (200-1400 ms) values.
• Acquisition of PDFF and T1 maps across four centers and vendors using 1.5T and 3T scanners with specific MRI sequences (SGRE for PDFF, MOLLI for T1).
• Quantitative analysis of maps with assessment of phantom temporal stability over 9 months using intraclass correlation coefficients (ICC) and reproducibility coefficients (RDC).

Main Results:

• PDFF mapping demonstrated excellent overall reproducibility (ICC=0.987, RDC=3.7%), with slight variability increase at higher T1 values.
• T1 mapping showed good reproducibility without fat (RDC 16-161 ms) but moderate to poor reproducibility with increasing fat content (RDC up to 1553 ms at 30% PDFF).
• Excellent temporal stability was observed for both PDFF and T1 mapping over 9 months.

Conclusions:

• This study confirms high reproducibility of PDFF mapping in liver disease assessment.
• T1 mapping reproducibility is significantly affected by fat and T1 levels.
• Fat-corrected T1 mapping is essential for reliable assessment of fibroinflammatory changes in chronic liver disease.