Access Type

Open Access Dissertation

Date of Award

January 2018

Degree Type


Degree Name



Physics and Astronomy

First Advisor

Yu-Chung N. Cheng


The first part of this thesis was to utilize existing methods to remove the unwanted background phase from phantom studies. Magnetic moments of straws filled by three different materials (gadolinium, ferritin, and nanoparticles) with four different concentrations of each material were subsequently quantified at 10 different echo times, with straws perpendicular to the main magnetic field. As the radius of each straw was known, susceptibility was calculated from each measured magnetic moment. Results from different echo times agreed within uncertainties. In addition, near perfect linear relation between quantified susceptibility and concentration was obtained for each material. However, phase values inside straws of nanoparticles and ferritin at certain concentrations appeared to deviate from theoretical values at various echo times. Nonetheless, the susceptibility and the additional frequency shift of each of those concentrations could be quantified from phase values inside straws obtained at perpendicular and parallel orientation. Such frequency shift may lead to incorrect susceptibility quantification when phase values inside the materials are used.

The second part of this thesis was to develop a procedure for properly removing the unwanted background phase. A reference phantom was utilized to remove the eddy current effect appearing in the target of interest. The susceptibility of the target, relative to that of air, was quantified through a least-squares fit with forward calculations of the induced magnetic fields based on segmented geometry of the target and the Green's function. The effects from shimming, imaging orientations, volume segmentation, partial volume effects, eddy current effects, distortions, fitting models, fitting regions, and errors were systematically investigated. Numerical simulations and phantom studies were performed to improve the fitting model and the procedure. An in vivo human head example was also presented to demonstrate the potential feasibility of the proposed method in clinical applications. By comparing the proposed method with other commonly used methods, the proposed method can properly remove most unwanted background phase in susceptibility weighted images with minimal overcorrections. The remaining phase reflects the actual phase distributions from the local structures of interest.

Included in

Physics Commons