Fat quantification using three-point dixon technique: in vitro validation.
RATIONALE AND OBJECTIVES: To test the repeatability, reproducibility and accuracy of the three-point Dixon (3PD) sequence for estimating true fat volume ratios using a fat/water phantom. MATERIALS AND METHODS: A phantom, constructed from test tubes of varying fat content, was imaged using the 3PD sequence on a 1.5T MRI scanner by two operators four times each. Fat volume ratios were calculated from these images and compared with true fat volumes. RESULTS: Measures of fat volume ratios calculated from the 3PD MR images correlated strongly with values for true fat volumes (r = 0.96). CONCLUSION: The 3PD technique was found to be highly reproducible and accurate, and may be useful for in vivo quantification of fat in lean tissues, such as the liver, pancreas or skeletal muscle.