Generalized k-space decomposition with chemical shift correction for non-Cartesian water-fat imaging

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Physics
• Biomedical Engineering

Background:

• Chemical-shift artifacts in non-Cartesian MRI are complex and clinically challenging.
• Conventional methods struggle with accurate water-fat decomposition and distortion correction.

Purpose of the Study:

• To introduce a novel k-space based iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL) approach.
• To simultaneously decompose multiple species and correct off-resonant species distortion in MRI.
• To improve water-fat separation and reduce chemical-shift artifacts in various MRI sequences.

Main Methods:

• Developed a new signal model accounting for phase accumulation in off-resonant spins across k-space trajectories.
• Adjusted the decomposition matrix for each k-space point during IDEAL processing.
• Demonstrated water-fat decomposition using projection reconstruction (PR)/radial, spiral, and Cartesian spin-warp imaging.
• Introduced an improved multipeak decomposition model for species with multiple resonant peaks.

Main Results:

• Achieved substantial correction of chemical-shift artifacts in phantoms and human subjects across PR/radial, spiral, and Cartesian spin-warp imaging.
• Simulations confirmed the nature of chemical-shift distortion for different acquisition types.
• The multipeak decomposition model showed substantial improvement in water-fat separation compared to single-peak approximations.

Conclusions:

• The novel k-space based IDEAL approach effectively corrects chemical-shift artifacts in non-Cartesian MRI.
• The technique offers improved water-fat separation and distortion correction with minimal increase in reconstruction time.
• The multipeak decomposition model enhances accuracy for complex chemical species like fat.