Author: Petr Bruza, Megan Clark, David J. Gladstone, Joseph Harms, Anthony E. Mascia, Brian W Pogue, Roman Vasyltsiv, Zhiyan Xiao, Rongxiao Zhang, Yongbin Zhang 👨🔬
Affiliation: Cincinnati Children's Proton Therapy Center, University of Missouri, Dartmouth College, Thayer School of Engineering, Dartmouth College, Washington University in St. Louis, University of Wisconsin - Madison 🌍
Purpose: With developing FLASH trials requiring diligent delivery monitoring, current proton therapy in-vivo dosimetry (IVD) cannot provide comprehensive verification, including dose and dose-rate volume histogram metrics, and is often restricted to 1D measurements which omit information about complex fields. This work presents a large-area time-resolved scintillator array approach to surface dose and dose-rate monitoring, which is projected across the patient volume to generate dose and dose-rate volume histograms.
Methods: Novel pixelated scintillator array was attached along an anthropomorphic phantom surface and imaged at 2500FPS during 250MeV UHDR proton irradiation. Stereovision modules captured the 3D shape of the array at 30FPS, allowing for dynamic angular optical emission correction and localization in the room coordinate-space. Surface measurements were validated against film, planned positions, and TPS-interpolated dose-rate. To emulate clinical workflow, the array was positioned at the phantom's rib region, CT-imaged, and mock treatment plan was delivered. Dynamic surface dose measurements were registered to CT and propagated through the volume using analytical depth projection to calculate dose and dose-rate volume histograms for lung, rib, and full intersected volumes.
Results: Cumulative surface dose achieved 99.6% gamma passing rate (2%/2mm) compared to delivery onto flat film and spot localization showed 0.32mm error. Surface dose-rate area histogram demonstrated 99.3% agreement with plan at 40Gy/s. Depth projected volume analysis showed agreement (8.21+/-0.13Gy) with log-derived dose at isocenter (8.1Gy), and <1% deviation in dose-rate volume coverage at 48Gy/s.
Conclusion: The high-speed surface dosimetry system presented in this work provides the first and only method of surface dose and dose-rate verification for UHDR proton therapy. The system enables both surface guidance and in-vivo dosimetry referenced to patient geometry independent of gantry coordinates. Surface dose dynamics are extrapolated through CT volume to derive dose and dose-rate volume histograms using intra-treatment data which is not achievable by any other IVD system.