Room: AAPM ePoster Library
Purpose: To evaluate surface-based target localization accuracy for spot scanning proton treatments in a 360° gantry.
Methods: Surface measurements were carried out using a modified version of the isocube phantom, which provided improved detectability in all planes. A CT simulation of the phantom was performed to make available the external contour as a localization surface. A plan was created placing isocenter on the central sphere of the phantom. With the phantom at treatment position, proton radiographs were acquired using a 2D ion-chamber array. The images were generated using a uniform field of 10.5 x 10.5 cm² with 0.5 cm spot spacing and a nominal energy of 156.6 MeV and a 40 mm range shifter. Prior to acquiring proton radiographs, the standard commissioning tests recommended by TG-147 were performed, i.e.: determination of field of view (FOV), spatial reproducibility, spatial drift, and, static and dynamic localization accuracy.
Results: After introducing a known shift of 1 mm to the phantom position based on surface imaging, the gamma passing rate at ?=1.0%/0.5 mm of the 2D-IC array dropped from 99 % to 76 %. Spatial drift and reproducibility showed values < 1.0 mm. The FOV available for accurate localization was found to be 12.5% less in comparison to the area available for scanning in the beam’s eye view (BEV), static and dynamic localization accuracy showed values = 1.0 mm.
Conclusion: Proton radiography provides a fast and reliable method to associate the information provided by a surface-imaging system and the irradiation field of the proton machine. Nevertheless, our commissioning results show that due to FOV limitations, surface-based localization in a 360° proton gantry has reduced capabilities. Therefore, inter and intra-fraction motion tracking is only available for certain patients in relative mode, i.e., acquiring a reference surface image after x-ray based IGRT has been performed.
Not Applicable / None Entered.
Not Applicable / None Entered.