ESTRO 2022 - Abstract Book

S195

Abstract book

ESTRO 2022

assurance (QA) recommendations [1]. A user-friendly QA tool was developed which through movement triggers a gating system and thereby measure temporal accuracy with high resolution in real-time. The aim of this work was to quantify the temporal accuracy for gated RT for six linacs and two different energies. Materials and Methods An in-house developed QA tool was used to measure the beam-on/off times for six beam-matched linacs (TrueBeam 2.5, Varian). Ten measurements per beam-on/off were performed for both 6 and 10 MV triggered by surface imaging (Catalyst, C-rad Positioning AB) clinically used for gated RT treatments. For all measurements, the integration time was set to 2000 µ s and a circular region-of-interest (ROI) with a radius of 20 mm was projected on a motion platform (fig 1b) to track the motion. The QA tool is equipped with three sources for the temporal accuracy measurements; two diodes (QED, Sun Nuclear) placed in the radiation field and a connection to the target current signal of the linac. The design of the electrical circuit is described elsewhere [2]. Each beam pulse generated is amplified and then measured with a high time resolution microcontroller unit. The first diode was used for all measurements presented, and the other two sources were used for verification only. The beam-on/off times (t on /t off ) were defined as the time between the start signal of the motion platform until the first/last beam pulse measured by the diode, corrected for a 22 ms time lag (fig 1a). Unpaired Mann-Whitney U- tests were carried out to compare t on and t off between the two energies.

Figure 1 . (a) Flow charts of the beam-on/off temporal accuracy measurements. (b) The motion platform of the QA tool with the ROI projected on top together with the diodes. Results The temporal accuracy for each linac is presented in figure 2. For all 60 measurements, the median t on and t off were 158 and 89 ms for both 6 and 10 MV (p = 0.68 and 0.73), with a large range (fig 2). Excellent agreement was observed for the diodes and target current signal, verifying the results of the primary diode. Approximately 20% of the t off values exceeded the 100 ms recommendation with a maximum value of 140 ms. For a dose rate of 600 MU/min this would in worst-case result in 1.4 MU delivered outside the gating window.