Related Experiment Video
Updated: Oct 4, 2025
08:51
Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
Published on: February 19, 2021
9.3K
Addressing concomitant gradient phase errors in time-interleaved chemical shift-encoded MRI fat fraction and R2 *
Nathan T Roberts1,2, Diego Hernando1,2,3,4, Nikolaos Panagiotopoulos1
1Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Magnetic Resonance in Medicine
|February 5, 2022
Summary
This study introduces a new method to correct phase errors in MRI scans caused by concomitant gradients. The technique improves the accuracy of fat fraction and R2* measurements, especially away from the scan
Area of Science:
- Magnetic Resonance Imaging (MRI)
- Medical Physics
- Biomedical Engineering
Background:
- Concomitant magnetic field gradients introduce phase errors in MRI, increasing with distance from the isocenter.
- These errors significantly affect quantitative MRI parameters like proton density fat fraction (PDFF) and R2*.
- Accurate estimation of PDFF and R2* is crucial for various clinical applications, including liver disease assessment.
Purpose of the Study:
- To develop and validate a novel complex-based fitting method to address concomitant gradient-induced phase errors in Chemical Shift Encoded (CSE) MRI.
- To enable accurate joint estimation of PDFF and R2* by incorporating pass-specific phase terms in multi-pass acquisitions.
- To provide a method that does not require prior knowledge of gradient waveforms.
Main Methods:
- A CSE-MRI spoiled gradient echo signal model was developed, incorporating pass-specific phase terms.
- Non-linear least squares estimation was employed for PDFF and R2* quantification.
- Cramér-Rao lower bound analysis was performed to assess noise performance, followed by validation in phantom and in vivo experiments.
Main Results:
- The proposed method effectively removed PDFF and R2* estimation errors up to 12% and 10 s⁻¹, respectively, at ±12 cm off-isocenter in phantom studies.
- In healthy volunteers, PDFF bias was reduced by ~10% and R2* bias by ~30 s⁻¹ at clinically relevant off-isocenter positions.
- Analysis of 29 liver datasets showed improved PDFF quantification with reduced bias and variability (8.4% improvement in coefficient of variation).
Conclusions:
- Concomitant gradient phase errors in multi-pass CSE-MRI lead to significant PDFF and R2* estimation biases, particularly away from the isocenter.
- The proposed complex-based fitting method accurately quantifies PDFF and R2* in the presence of these phase errors.
- This approach offers a robust solution for quantitative MRI without needing gradient waveform information.
