Keywords:
Performed at one institution, Experimental, Not applicable, Image verification, Artifacts, Technical aspects, Radiation therapy / Oncology, Experimental investigations, MR, Oncology, MR physics, MRI
Authors:
Y. kato, K. okudaira, M. kumagai, T. Taoka, T. Kamomae, Y. Itoh, S. Naganawa; Nagoya/JP
DOI:
10.26044/ecr2020/C-01229
Methods and materials
All scans were performed on a 3-T MR scanner (MAGNETOM Skyra, Siemens, Erlangen, Germany) using a body coil. We used a 200 mm diameter pin phantom (90–401 type; Nikko Fines, Japan) to cover the maximum 500 mm FOV of the scanner. Consequently, we performed phantom scans at nine separate positions (three positions for in-plane direction by three positions for through-plane direction with 80 mm overlap [see Fig. 3]) using a MR-certified laser bridge (DORADOnova MR3T; LAP, Germany) for alignment purposes. Images were obtained following four sequences including 2D T1-weighted spin echo (SE), 2D T2-weighted fast spin echo (FSE), 3D T1-weighted gradient echo (GRE) and 3D T2-weighted FSE by axial plane and coronal plane. All images were scanned without DC, and vendor specific 2D- and 3D-DC were applied by retrospective reconstruction on the scanner’s console. Table 1 summarizes the detailed scan parameters for all sequences used in this study. Images of nine positions were further combined with direct coronal plane, and multi-planar reconstruction from transverse to coronal plane. Furthermore, the distance from the isocenter to each control point was measured, and the measured distributions were compared with the theoretical ones (Fig. 4). The distortion was calculated using the following formula:
Distorsion = √(XMRI − XTheory)2 + (ZMRI − ZTheory)2 (1)
The in-plane and through-plane geometric distortions were quantitatively evaluated for no-DC, 2D-DC, and 3D-DC.