Accelerated T2*-compensated fat fraction quantification using a joint parallel imaging and compressed sensing...

Area of Science:

• Magnetic Resonance Imaging
• Medical Imaging Physics

Background:

• Accurate water-fat separation is crucial for quantitative magnetic resonance imaging (MRI).
• Existing methods may be limited in acceleration capabilities.
• T2* decay can affect the accuracy of fat fraction quantification.

Purpose of the Study:

• To develop a T2*-compensated parallel imaging and compressed sensing framework for water-fat separation.
• To enable accelerated quantitative imaging of proton density fat fraction.

Main Methods:

• A two-stage estimation process was developed, first approximating the B0 field map.
• The method jointly estimates and refines R2* and B0 field maps, compensating for T2* decay.
• Validation was performed using water-fat phantoms and liver MRI datasets from adult volunteers.

Main Results:

• The T2*-compensated method yielded accurate fat fraction estimates in phantoms.
• Liver datasets showed accurate fat fraction estimation at acceleration factors up to 4×.
• The proposed method achieved higher acceleration factors compared to sequential methods, despite minor artifacts.

Conclusions:

• The developed framework effectively compensates for T2* decay in water-fat imaging.
• This T2*-compensated approach shows potential for further accelerating water-fat imaging.
• Accurate proton density fat fraction quantification can be maintained at increased speeds.