Liver fat quantification using a multi-step adaptive fitting approach with multi-echo GRE imaging.

Area of Science:

• Medical Imaging
• Quantitative MRI
• Biophysics

Background:

• Accurate quantification of liver fat is crucial for diagnosing and managing hepatic steatosis.
• Existing methods for measuring proton density fat fraction (PDFF) and R(2)* may have limitations in accuracy and robustness.

Purpose of the Study:

• To develop and validate a novel multi-step adaptive fitting approach for liver PDFF and R(2)* quantification.
• To assess the performance of this method on a widely accessible hardware platform.

Main Methods:

• Utilized a multi-echo 3D gradient echo sequence.
• Employed Dixon decomposition for initial fat/water fraction estimation.
• Applied a multi-step nonlinear fitting procedure with a multi-peak fat spectral model to refine PDFF and R(2)* values.

Main Results:

• The method demonstrated strong agreement with numerical phantom ground truth.
• Results were robust across variations in field strength, homogeneity, readout type, SNR, and echo times.
• In vivo validation showed excellent consistency between PDFF measurements and spectroscopy.

Conclusions:

• The developed multi-step adaptive fitting approach is effective in both simulated and initial clinical evaluations.
• This method holds significant potential for improving the quantification of hepatic steatosis.