Room: Exhibit Hall
Purpose: To reconstruct dose delivered to a lung tumor from a passive proton beam, the kinematics of proton delivery was investigated. The effect of the variation of ventilator period, the magnitude of target motion, and periodic dose rate on the delivered dose was investigated.
Methods: We employed an in-house 4D lung phantom that consists of an acrylic cylinder embedded in cork. The amplitude of 2 cm and a breathing period of 6 seconds were used. A 4D CT of the phantom was obtained with 10 phases and a treatment plan was generated on the average axial (AVEA) CT using maximum intensity projection (MIP) for target contouring. Migrating a fixed AVEA plan to each of the 10 phase CT volume-set, re-calculation of dose distribution was done for generation of temporal dose distribution. A dose rate was modeled using a periodic spill structure of 2.2 sec/cycle with measured accelerator spill duration of 84.5ms. By fixing phase zero time to start of spill structure, reconstruction of the 3D dose was obtained for two distinct fractionation schemes of conventional treatment and SBRT. 2D dose distribution for each treatment scheme at the center of the target is compared to corresponding AVEA distribution via gamma index.
Results: The reconstructed 2D dose based on two-fractionation schemes exhibit a maximum variation of up to 1.8% compared to AVEA dose. The dose discrepancy for SBRT reconstructed dose was even smaller. Reconstructed 2D dose distribution displayed a dependence on the ratio of the number of incomplete to complete ventilator periods.
Conclusion: Reconstruction of 3D Dose distribution in proton lung treatment is dependent on respiratory cycle and also on its treatment scheme. The best agreement between AVEA dose delivery and the temporal delivered dose is when the temporal dose dependence is washed out, e.g., for large MU treatments