ESTRO 2020 Abstract book

S387 ESTRO 2020

OC-0636 Highly spatially resolved 2D plastic scintillation detector system for small field real-time QA P. Pittet 1 , J. Ribouton 2 , P. Jalade 2 , J. Esteves 1 , F. Blanc 3 , G. Haefeli 3 , P. Hopchev 3 , J. Galvan 4 , G. Lu 1 1 Institut des Nanotechnologies de Lyon, Université Claude Bernard Lyon 1, Villeurbanne, France ; 2 Hospices Civils de Lyon, Centre Hospitalier Lyon Sud, Pierre Bénite, France ; 3 EPFL, Laboratoire de Physique des Hautes Energies LPHE, Lausanne, Switzerland ; 4 Institut des Nanotechnologies de Lyon, CPE Lyon, Villeurbanne, France Purpose or Objective Beam characterization during commissioning and periodic QA verifications for SRS treatment equipment relies nowadays on time consuming procedures. To address this problem, we propose the use of the SciFi detector, a real- time 2D detector (initially developed for the LHCb experiment at CERN) for commissionning and QA of SRS systems. Material and Methods The SciFi detector is made of tissue-equivalent scintillating plastic fibers arranged in a staggered geometry to form a ribbon with a fiber pitch of 275μm as shown in Fig. 1. For TMR measurements, a single SciFi ribbon is placed in a RW3 solid-water plate arrangement and is irradiated from the side of the ribbon. For small field monitoring, up to 6 SciFi ribbons are stacked at the depth of maximum dose in a RW3 phantom with in-plane 30° tilted orientations. The scintillating signal at the detector output which represents the projected irradiation profile along the fibers axis is acquired by either a sCMOS camera or Si phototodiode linear arrays (with an element pitch of 400µm). The acquired images allow an iterative tomographic reconstruction of the field for QA procedures. The detector system has been tested with a 6MV photon beam for measuring TMR curves on the one hand and for characterization of SRS cone collimators on the other.

measurements (within ±0.32mm for both FWHM and penumbra region). The system can thus serve to verify the reproducibility of conical collimator insertion into mount. Measured output factors (OF) with the proposed approach are within ±1.6% for all the tested cone sizes as compared with OF determined independently by IRSN.

Conclusion We have tested a SciFi plastic scintillating detector for beam monitoring during QA and commissioning procedures of SRS treatment. It allows implementation of accurate and time-efficient QA procedures for determining TMR curves, FWHM, penumbra and OF for fields based on SRS cone collimators. As the SciFi detector is tissue equivalent, it could be implemented for full End-To-End testing of SRS treatment. OC-0637 Identifying links between beam modeling accuracy and IROC phantom performance M. Glenn 1 , R. Howell 1 , J. Pollard-Larkin 1 , C. Peterson 2 , D. Followill 1 , S. Kry 1 1 The University of Texas MD Anderson Cancer Center, Radiation Physics, Houston, USA ; 2 The University of Texas MD Anderson Cancer Center, Biostatistics, Houston, USA Purpose or Objective Several beam modeling parameters such as the dosimetric leaf gap (DLG) and multileaf collimator (MLC) transmission have been previously identified as important factors for treatment planning system (TPS) dose calculation accuracy. Due to the large presence of dose calculation errors in IROC Houston anthropomorphic phantom irradiations, this work seeks to identify links between beam modeling parameter choices and subsequent phantom performance, based on self-reported TPS beam modeling data. Material and Methods 781 phantoms performed between August 2017 and July 2019 were reviewed. Individual cases were evaluated and excluded from analyses if it was determined that phantom performance was not directly related to dose calculation disagreement (i.e. incorrect treatment setup). Poorly- performed irradiations were defined as those having dose deviations in excess of 5% when compared to the TPS- calculated dose. Self-reported beam modeling values from each selected irradiation were then compared with previous IROC survey results, which describe consensus values for parameters modeling the MLC and spot size (Table 1). Cases were flagged when any reported parameter was outside the 10 th and 90 th percentiles of IROC survey data.

Results The sCMOS camera is suitable for high resolution readout of the SciFi detector (40µm/pixel), but requires 20MU to acquire images for fields as small as 4mm. The readout based on the photodiode array provides a much higher sensitivity (a few MU are sufficient for a good signal to noise ratio) with a lower 400µm resolution. The SciFi detector system used for simultaneous measurement of the on-axis full TMR curve has a high resolution in depth. The TMR curve shown in Fig. 2 is measured within 2s, which is much shorter than for 3-axis water scanning phantom based procedures. Moreover, the obtained curve is particularly well defined in the build-up region due to the 275µm fiber pitch. The SciFi detector system is also used for SRS field monitoring for 4, 5, 6, 7.5, 10, 12.5 and 15mm cone collimators: results are in good agreement with EBT3 film