ESTRO 2025 - Abstract Book

S3707

Physics - Quality assurance and auditing

ESTRO 2025

small effect to the image quality when carried out with an anthropomorphic phantom. As a result, it becomes crucial and possible to establish dose thresholds for clinically used protocols in order to safeguard patients undergoing radiation therapy.

Keywords: Cone-Beam CT, optimization, dose

References: Elekta Medical Linear Accelerator XVI R5.0.6, Instructions for use for: Elekta Synergy®, Elekta Axesse™, Elekta Infinity™, VersaHD™

4581

Proffered Paper Remote Dose Rate Assessment for FLASH Quality Assurance Hayden Scott 1 , Paige Taylot 1 , Uwe Titt 1 , Michele Kim 2 , Ryan Sun 3 , Alexander Baikalov 1 , Luke Connell 1 , Brett Velasquez 1 , Nolan Esplen 1 , Stephen Kry 1 , Emil Schueler 1 1 Radiation Physics, MD Anderson, Houston, USA. 2 Radiation Oncology, University of Pennsylvania, Philadelphia, USA. 3 Biostatistics, MD Anderson, Houston, USA Purpose/Objective: To develop a remote, portable audit system that can capture the temporal information for ultra-high dose rate proton and electron beams as used in FLASH. Specifically, this system needed to be capable of broadly measuring dose-rate information across various delivery systems, enabling the Imaging and Radiation Oncology Core (IROC) to conduct remote audits in support of clinical trials. Material/Methods: Semiconductor-based detectors were evaluated for their ability to measure temporal information 1 . Specifically, a PTW flashDiamond, an Exradin D1H diode, an FD11A silicone photodiode, and a FGA01 InGaAs photodiode were evaluated. All detectors were terminated with a 50 Ohm resistor and operated from a microprocessor (Red Pitaya StemLab124-15). Protons were delivered from a synchrotron in a single spill to a dose of 4 Gy, 8 Gy, and 20 Gy to the detectors (200 Gy/s). Matched doses were also delivered using the 9 MeV electron beam from a FLASH Mobetron using two different dose-rate settings (1 Gy/pulse, 83 Gy/s and 4 Gy/pulse, 300 Gy/s). The microprocessor was operated between 125 MHz and 61KHz to acquire data in a single 16,384 sample buffer depending on the time scale of the beam delivery. The measured temporal information compared to established standards; For the proton deliveries, spill length (t) was compared to an Advanced Markus Chamber . For the electron deliveries, the pulse length (t) and time between pulses (T) were compared to a reference Beam Current Transformer. Results: Timing information measured with the different remote detectors evaluated is shown in Table 1, and representative examples of the remote-capable measurement versus the established standard are shown in Figure 1. The flashDiamond and D1H measured the timing of proton spills and single electron pulses with error less than 1.5% with a mean error of 0.6 +/- 0.5%. Photodiodes measured single proton spills and single electron pulses with an average percent error of 2.0 +/- 1.6 due to slower fall-off times. Single electron pulses had percent error of 4.7% and over 300% for the FGA01 and FD11A respectively. For multi-pulsed electron beams, all detectors were able to measure the inter-pulse timing within 1% mean error with an average error of 0.5 +/- 0.3%.