Isotropic Three-Dimensional MRI-Fricke-Infused Gel Dosimetry

博士 === 國立陽明大學 === 醫學工程研究所 === 101 === Fricke-infused gel has been shown to be a simple and attainable method for the conformal measurement of absorbed radiation dose. Nevertheless, its accuracy is seriously hindered by the irreversible ferric ion diffusion during magnetic resonance imaging (MRI)...

Full description

Bibliographic Details
Main Authors: Nai-Yu Cho, 卓乃瑜
Other Authors: Woei-Chyn Chu
Format: Others
Language:en_US
Published: 2013
Online Access:http://ndltd.ncl.edu.tw/handle/35792631668631042188
Description
Summary:博士 === 國立陽明大學 === 醫學工程研究所 === 101 === Fricke-infused gel has been shown to be a simple and attainable method for the conformal measurement of absorbed radiation dose. Nevertheless, its accuracy is seriously hindered by the irreversible ferric ion diffusion during magnetic resonance imaging (MRI), particularly when three-dimensional (3D) dose measurement in stereotactic radiosurgery is considered. In this study, we developed a fast three-dimensional spin-echo based Fricke gel dosimetry technique to reduce the adverse effects of ferric ion diffusion and to obtain an accurate isotropic 3D dose measurement. This thesis presents the investigation of the fast MRI-Fricke-infused gel dosimetry in two parts. In the first part, a skull shaped phantom containing Fricke-infused gel was irradiated using Leksell Gamma Knife. The rapid image-based dosimetry technique was applied with the use of a 3D fast spin-echo (3D-FSE) magnetic resonance imaging sequence. We mathematically derived and experimentally validated the correlations between dose-response characteristics and parameters of the 3D fast spin-echo MR imaging sequence. Absorbed dose profiles were assessed and compared to the calculated profiles given by the Gamma Knife treatment planning system. Coefficient of variance (CV%) and coefficient of determination (R^2) were used to evaluate the precision of dose-response curve estimation. The agreement between the measured and the planned 3D dose distributions was quantified by gamma-index analysis of two acceptance criteria. Film dosimetry was also performed to verify the accuracy of dose distributions from Gamma Knife treatment planning system. In the second part, Fricke-infused gel cube inside the head and neck anthropomorphic phantom was irradiated by CyberKnife system, and the subsequent absorbed dose evaluations were performed using the proposed protocol in the first part. Two dose distributions from the gel measurement and CyberKnife planning system have been compared. Proper magnetic resonance imaging parameters were explored to render an accurate three-dimensional absorbed dose mapping with a 1 mm^3 isotropic image resolution. The efficacy of the dose-response estimation was approved by an R^2 > 0.99 and an average CV% of 1.6%. Average gamma pass-rate between the experimentally measured and GammaPlan calculated dose distributions were 83.8% and 99.7% for 2%/2mm and 3%/3mm criteria, respectively. Satisfying results have also been obtained in the CyberKnife case. With the designed MR imaging sequence and parameters, total 3D MR acquisition time was confined to within 20 minutes post-irradiation, during which time ferric ion diffusion effects were negligible, thus enabling an accurate 3D radiation dose measurement.