Related Experiment Video
Updated: Sep 16, 2025

Fat-Water Phantoms for Magnetic Resonance Imaging Validation: A Flexible and Scalable Protocol
Published on: September 7, 2018
Motion-Insensitive Flip Angle Modulated Liver Proton Density Fat-Fraction and R2* Mapping During Free-Breathing MRI
Julius F Heidenreich1,2, Jiayi Tang1,3, Daiki Tamada1
1Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Background:
Breath-held quantitative three-dimensional chemical-shift-encoded (3D-CSE)-MRI is widely used for liver fat and iron quantification but is limited in patients unable to hold breath. Motion-insensitive two-dimensional flip angle modulated (2D-FAM) acquisitions with short temporal aperture may address this limitation.
Purpose:
To evaluate the performance of 2D-FAM-based CSE-MRI during free-breathing for quantifying proton density fat-fraction (PDFF) and R2*-based liver iron concentration (LIC).
Study Type:
Retrospective single-center observational analysis.
Population:
230 patients (110 women [47.8%], mean age 51 ± 16 years [standard deviation, SD], range 4-85), undergoing routine clinical care between January and August 2024.
Field Strength/Sequence:
1.5-T and 3.0-T, comparison of 2D-FAM-CSE-MRI under free-breathing and breath-held 3D-CSE-MRI.
Assessment:
Three radiologists evaluated image quality, apparent signal-to-noise ratio (SNR), and motion artifacts on a 5-point scale. PDFF and R2*-based LIC were measured in each liver segment from each acquisition. Agreement of PDFF and LIC between techniques was assessed, comparing body mass index (BMI) groups (< 30 and ≥ 30 kg/m2). Inter-rater reliability was assessed for all analyses.
Statistical Tests:
Inter-rater reliability with intra-class coefficient (ICC). Bias and limits of agreement (LoA) using Bland-Altman analysis. Linear regression assessed correlations of PDFF and LIC. Receiver operating characteristic (ROC) curves (AUC) to assess diagnostic performance. Statistical significance was set at p < 0.05 (Bonferroni-corrected p < 0.01 for patient characteristics).
Results:
2D-FAM demonstrated superior image quality (median rating 4 vs. 3) and fewer motion artifacts (median rating 4 vs. 3) than 3D-CSE-MRI in PDFF and R2* maps. Inter-rater reliability was good to excellent (ICC 0.77-0.92). PDFF agreement was excellent (bias: +0.32%, LoA [-1.37,2.00] for BMI < 30 kg/m2; +0.66%, [-2.44,3.76] for BMI ≥ 30 kg/m2). LIC agreement was high (bias: -0.02 mg/g, [-0.22,0.18]) up to a threshold of 7.0 mg/g. ROC analysis showed AUC up to 1.00 for PDFF and LIC thresholds.
Data Conclusion:
2D-FAM CSE-MRI enables motion-insensitive liver fat and iron quantification that can be performed during free-breathing.
Evidence Level:
2.
Technical Efficacy:
Stage 2.
Related Concept Videos
Magnetic Resonance Imaging
Imaging Studies for Cardiovascular System IV: CMRI

