Development of a physical geometric phantom for geometric and dosimetric deformable image registration accuracy evaluation

概要

（書式１８）

学
（

位 論 文 要 約
A b s t r a c t ）

博士論文題目 Title of dissertation
Development of a physical geometric phantom for geometric and dosimetric deformable image registration accuracy
evaluation
(非剛体レジストレーションの幾何学的及び線量的評価を目的とした物理ファントムの開発)

東北大学大学院 医学系研究科 医科学専攻
内科病態学講座 放射線腫瘍学分野
Siwaporn Sakulsingharoj
Purpose: The purposes of this study were to develop a physical geometric phantom for deformable image
registration (DIR) accuracy evaluation, to evaluate geometric DIR accuracy using various custom inserts and to evaluate
dosimetric DIR accuracy using radiophotoluminescent glass dosimeter (RPLD) and custom developed inserts in which
the point dosimeters can be placed.
Methods and materials: The phantom reproduced tumor shrinkage, rectum shape change, and body shrinkage using
several physical phantoms with custom inserts. I tested the feasibility of the proposed phantom using 5 DIR patterns at
17 domestic and 2 international institutions (21 datasets). Eight institutions used the MIM software, seven used Velocity,
and six used RayStation. The geometric DIR accuracy was evaluated using the Dice similarity coefficient (DSC) and
Hausdorff distance (HD). Furthermore, the phantom was developed to facilitate simultaneous evaluation of geometric
and dosimetric accuracy of DIR. Four computed tomography (CT) images of the phantom were acquired with four
different settings. Four volumetric modulated arc therapy (VMAT) plans were computed for different phantom. Two
different patterns were applied to combination of four phantom settings. RPLD dose measurement was combined
between corresponding two phantom settings. DIR-based dose accumulation was calculated between corresponding two
CT images with two commercial DIR software and various DIR parameter settings. Accumulated dose calculated using
DIR was then compared with measured dose using RPLD.
Results: The mean ± standard deviation (SD) of DSC were 0.909 ± 0.088 (range, 0.434–0.984) for tumor proxy and
0.909 ± 0.048 (range, 0.726–0.972) for rectum proxy. The mean ± SD of HD were 5.02 ± 3.32 mm (range, 1.53–20.35
mm) for tumor proxy and 5.79 ± 3.47 mm (range, 1.22–21.48 mm) for rectum proxy. In three patterns evaluating the DIR
1

（書式１８）
accuracy within the entire phantom, 61.9% of the data had more than a DSC of 0.8 in both tumor and rectum proxies. In
two patterns evaluating the DIR accuracy by focusing on tumor and rectum proxies, all data had more than a DSC of 0.8
in both tumor and rectum proxies. The mean ± SD of dose difference was 2.71 ± 0.24% (range, 2.22–3.01%) for tumor
proxy and 3.79 ± 0.80% (range, 1.56–4.83%) for rectum proxy. The mean ± SD of target registration error (TRE) was
1.75 ± 1.39 mm (range, 0.03–4.43 mm) for tumor proxy and 7.02 ± 5.59 mm (range, 0.54–17.47 mm) for rectum proxy.
These results suggested that DIR accuracy had wide range among DIR parameter setting.
Conclusions: The wide range of DIR performance highlights the importance of optimizing the DIR process. Thus,
the proposed method has considerable potential as an evaluation tool for geometric DIR accuracy and quality assurance
(QA). The dose difference observed in my study was 3% for tumor proxy and within 5% for rectum proxy. The custom
developed physical phantom with inserts showed potential for accurate evaluation of DIR-based dose accumulation. The
prospect of simultaneous evaluation of geometric and dosimetric DIR accuracy in a single phantom may be useful for
validation of DIR for clinical use. ...