Understanding the morphological mismatch between magnetic susceptibility source and t2* image

Area of Science:

• Medical Imaging
• Biophysics
• Magnetic Resonance Imaging

Background:

• T2*-weighted magnetic resonance imaging (T2*MRI) is widely used, but its images do not directly represent the spatial distribution of magnetic susceptibility (χ) sources.
• Understanding the discrepancies between χ sources and their T2* images is crucial for accurate interpretation.

Purpose of the Study:

• To investigate the reasons for morphological mismatches between magnetic susceptibility sources and their T2*-weighted images.
• To numerically characterize these source/image mismatches using simulations.

Main Methods:

• Numerical simulations of T2*MRI were performed by defining 3D χ sources and calculating complex-valued T2* images.
• Intravoxel dephasing, with and without diffusion, was considered.
• An empirical α-power model was proposed to describe the source/image transformation.
• Morphological mismatches were quantified using spatial correlation and full width at half maximum (FWHM) for Gaussian-shaped χ sources.

Main Results:

• Theoretical analysis showed χ-induced fieldmaps differ morphologically from χ sources due to the dipole effect.
• T2* magnitude images undergo additional morphological changes due to a quadratic transformation of the fieldmap in the small phase angle regime.
• Numerical simulations indicated T2* magnitude images can exhibit 20%-25% morphological shrinkage compared to the χ source.
• T2* phase images closely resemble the fieldmap, thus retaining the dipole-induced morphological mismatch from the χ source.

Conclusions:

• Morphological mismatches between bulk χ sources and T2* images arise from 3D convolution (dipole effect), nonlinearities in T2* magnitude/phase calculations, and spin diffusion.
• In the small phase angle regime, T2* magnitude images show overall morphological shrinkage.
• T2* phase images are affected by the dipole effect but preserve the χ-induced fieldmap morphology.