Highly Reproducible, Vendor-Agnostic, Motion-Insensitive Liver PDFF Mapping at 0.55T, 1.5T, and 3T.

Area of Science:

• Medical Imaging
• Quantitative MRI
• Biomedical Engineering

Background:

• Proton-density fat-fraction (PDFF) quantification is crucial in various clinical applications.
• Existing methods can be vendor-specific and susceptible to motion artifacts.
• There is a need for a standardized, motion-insensitive PDFF quantification technique.

Purpose of the Study:

• To develop and validate a vendor-agnostic, motion-insensitive PDFF quantification method.
• To implement and test the Flip-Angle-Modulated (FAM) 2D chemical-shift-encoded (CSE) MRI technique.
• To ensure the method's accuracy and reproducibility across different MRI vendors and field strengths.

Main Methods:

• Developed a vendor-agnostic platform (Pulseq-FAM) and a vendor-specific platform (GE-specific FAM) for PDFF quantification.
• Distributed implementations across four sites with twelve MR systems from three vendors (Siemens, GE, Philips) at 0.55T, 1.5T, and 3T.
• Validated using a phantom with varying PDFF (0-30%) and T1 values, and conducted volunteer studies (adults and children) assessing free-breathing and breath-held scans.

Main Results:

• Pulseq-FAM demonstrated reduced T1-bias and between-system variability compared to 3D-CSE in phantom studies.
• Volunteer studies showed improved image quality and reduced motion artifacts with Pulseq-FAM versus 3D-CSE (p < 0.01).
• Excellent agreement was observed between Pulseq-FAM and 3D-CSE (LoA = 3.4% PDFF), and Pulseq-FAM and GE-specific FAM (LoA = 2.0%), with high repeatability and reproducibility.

Conclusions:

• Pulseq-FAM provides accurate, reproducible, vendor-agnostic, and motion-insensitive PDFF quantification.
• The method is effective for both adult and pediatric populations.
• This technique enhances the reliability of fat quantification in MRI across diverse clinical settings.