[Discussion and improvement methods of quantitative susceptibility mapping reconstruction]

Area of Science:

• Medical Imaging
• Biophysics
• Image Processing

Background:

• Quantitative Susceptibility Mapping (QSM) relies on accurate background field removal.
• Existing methods like SHARP, RESHARP, and LBV have limitations in speed, artifact control, and complexity.
• The truncated k-space division (TKD) method is prone to severe artifacts.

Purpose of the Study:

• To assess and compare common background field removal methods in QSM.
• To analyze the sources of artifacts in the TKD method.
• To propose and validate an improved TKD method for artifact suppression in susceptibility inversion.

Main Methods:

• Phase images were acquired using gradient echo sequences.
• Comparison of Sophisticated Harmonic Artifact Reduction for Phase data (SHARP), Regularization Enable of SHARP (RESHARP), and Laplacian Boundary Value (LBV) methods based on quality and speed.
• Analysis of TKD artifacts due to truncation and discontinuity, followed by proposing an improved TKD with adjusted truncation and enhanced continuity.
• Comparative susceptibility inversion using original and improved TKD methods.

Main Results:

• SHARP and RESHARP offered fast reconstruction, but SHARP had significant artifacts and low precision; RESHARP implementation was complex.
• LBV provided high precision and prominent image details but was slow.
• The original TKD method exhibited severe reconstruction artifacts.
• The improved TKD method demonstrated substantial artifact suppression and yielded good inversion results in artifact-prone regions.

Conclusions:

• No single existing background field removal method is optimal for all QSM applications.
• The proposed improved TKD method effectively mitigates artifacts inherent in the standard TKD approach.
• The enhanced TKD method shows promise for improving the accuracy and reliability of QSM.