Room: Davidson Ballroom A
Purpose: Recent advances in deformable dose accumulation software and motion management systems have started to change the landscape in adaptive radiotherapy. Validation of these systems using point or planar measurements in simple rigid motion phantoms is the most common method of commissioning, but this does not account for the impact of 3-D translational and deformable motion on the absorbed dose distribution. The goal of this work was to develop a deformable, anthropomorphic abdominal phantom which incorporates 3-D gel dosimetry for the validation of these systems.
Methods: Plastisol polyvinyl chloride (PVCP) was used to create a deformable abdominal phantom deformed by a 1-D motion stage. Organs at risk were represented by lower density PVCP and casting material. A cavity was created within the phantom to incorporate deformable nPAG polymer gel dosimeters contained within PVCP cups. The cup outer diameter matched the cavity, while the inner cavity containing nPAG gel was molded into an asymmetrical tumor shape with a 3-D printed insert. Three separate measurements of a liver SBRT treatment were performed using a section of each gel dosimeter as the target.
Results: The phantom allowed for 3 cm of deflection without damage. The average gamma pass rate of the gel dosimeter dose distributions were 84.6%Â±2.1% (3%/3 mm) and 97.0%Â±0.5 (3%/5 mm) using a 20% dose threshold. The effects of edge oxygen inhibition and image coregistration difficulties are apparent. This is demonstrated by limiting the region of interest to the central gel slices. The gamma pass rates in this region are 96.6%Â±1.8% (3%/3 mm) and 99.7%Â±0.5% (3%/5 mm).
Conclusion: A deformable anthropomorphic abdominal phantom has been developed for the validation of deformable dose accumulation algorithms and intrafractional motion management systems. An analysis of calculated and measured doses demonstrates the utility of this novel phantom for moving targets in radiotherapy.
Funding Support, Disclosures, and Conflict of Interest: This work was partially funded by NIH Grant R01CA190298.