Computed tomography images with different resolutions in Sentinel system:a setup error analysis / 中华放射肿瘤学杂志

ABSTRACT

Objective To study the impact of setup error caused by computed tomography ( CT) images with different resolutions in the Sentinel system on clinical treatment. Methods A phantom was scanned by large?aperture positioning CT with two different resolutions ( CT10. 5 mm × 0. 5 mm × 1. 0 mm, FOV 256 mm, Matrix 512, thickness 1 mm;CT31. 0 mm×1. 0 mm×3. 0 mm, FOV 500 mm, Matrix 512, thickness 3 mm) . The CT images were transferred to the planning system. The radiation fields were designed and transferred to MOSAIQ and Sentinel systems. Ten fixed setup errors were applied to a six degree of freedom couch. The Sentinel system was used to position the two groups of CT images and generate the setup errors. The comparison of two datasets was made by paired t?test. Cone?beam CT was used for independent verification. Results The setup errors in x?, y?, and z?directions were significantly smaller on CT1 than on CT3(0.19±0. 11 vs. 0.33±0. 16 mm, P=0. 061;0.59±0. 79 vs. 1.07±1. 09 mm, P=0. 008;0.67±0. 75 vs. 1.16±1. 30 mm, P=0. 043). There were no significant differences in rotational errors in x?, y?, or z?directions between the two datasets ( P=0. 494;P=0. 182;P=0. 298) . Conclusions The Sentinel system has a higher setup accuracy in the 0. 5 mm×0. 5 mm×1. 0 mm mode than in the 1. 0 mm×1. 0 mm×3. 0 mm mode. However, the later mode is still an acceptable choice in clinical treatment.