Identifier

etd-06182015-155915

Degree

Master of Science (MS)

Department

Physics and Astronomy

Document Type

Thesis

Abstract

Chemical exchange saturation transfer (CEST) and magnetization transfer (MT) are types of magnetic resonance imaging (MRI) experiments in which contrast is based on the transfer of magnetization from selectively saturated solute or macromolecular protons to bulk water protons. These processes offer insight into the chemical composition of tissue and are quantified by the asymmetry of the magnetization transfer ratio (MTRasym). This study was to develop a Z-spectral curve fitting procedure based on the underlying physics of CEST-MRI from which MTRasym values can be calculated and applied to distinguish healthy tissue from cancer. Z-spectra were collected from CEST-MR images of a phantom. The data were fit to both the proposed model which separately fits the upfield and downfield regions of the Z-spectra, and two polynomial models from literature. A preferred model was identified using the small sample bias-corrected Akaike’s Information Criterion (AICc). Z-spectra were collected from CEST-MR images of prostate cancer patients and fit with the same models; the preferred model was selected using the AICc. CEST-MR images of bladder cancer patients were acquired and the Z-spectra were fit with the preferred model identified from the phantom images. MTRasym was calculated at frequency offsets of 3.5 ppm and 2.0 ppm to determine if these quantities were capable of distinguishing normal bladder wall (NBW) from bladder cancer. The proposed fitting model with a 5th order polynomial for the downfield region was the preferred curve fitting model by the AICc model selection procedure for the phantom while a 6th order polynomial was preferred for the prostate cancer Z-spectra. MTRasym(2.0 ppm) values calculated from the bladder cancer Z-spectra did not differ significantly between the NBW and tumor regions. A statistically significant difference existed between the NBW and tumor regions for the MTRasym(3.5 ppm) values (p < 0.001). The proposed model was preferred to the polynomial models from literature based on the AICc metric. Application of the technique to patient images showed the potential to distinguish NBW from bladder cancer based on the statistically significant MTRasym(3.5 ppm) values in these regions.

Date

2015

Document Availability at the Time of Submission

Student has submitted appropriate documentation to restrict access to LSU for 365 days after which the document will be released for worldwide access.

Committee Chair

Jia, Guang

Share

COinS