ESTRO 2020 Abstract book

S751 ESTRO 2020

Conclusion The detector can be effectively used for small fields dosimetry in quality controls of MRgRT. Further studies, regarding the elaboration of 2D arrays and the integration into MR surface coils are currently under development to use this detector as in-vivo dosimeter during online MRgART PO-1330 On the Implementation and Validation of 3D Computational Pediatric Phantoms in Commercial TPS A. Gupta 1 , Y. Qiao 1 , S. Shrestha 1 , C. Owens 1 , C. Lee 2 , C. Ditty 3 , S. Smith 1 , R. Weathers 1 , R.M. Howell 1 1 University of Texas MD Anderson Cancer Center, Radiation Physics, Houston, USA ; 2 National Cancer Institute, Division of Cancer Epidemiology & Genetics, Bethesda, USA ; 3 RaySearch Laboratories, Physics, Cincinnati, USA Purpose or Objective Our group developed an in-house computational phantom that can be scaled to any age from infant to adult. The age-based scaling accounts for non-uniform growth of different body regions (in three dimensions [3D]). Our phantom was originally modeled in FORTRAN and has been used for more than three decades for many late effects studies of pediatric cohorts treated with conventional radiation therapy (RT). However, the phantom cannot be used for cohorts treated with modern RT with complex beam arrangements designed using computed tomography (CT) based treatment planning. Here, we aimed to (1) develop a 3D model of the phantom in Digital Imaging and Communications in Medicine (DICOM) format, (2) scale the phantoms’ body regions and organs to different pediatric ages, and (3) convert the phantom from FORTRAN to DICOM format across the full age range. Material and Methods We developed a script to generate the 3D model of the phantom in DICOM format using RayStation treatment planning system (TPS). A phantom of height 42.20 cm was modelled as a master phantom that can be scaled to any age from infant to adult using the non-uniform age-scaling growth functions and equations. These scaling functions were developed from data published by the Society of Automotive Engineers (Fig 1A-C), which included body dimension measurements of more than 4000 children ages 3 months, 1, 3, 5, 10, 15 and 18 years (adult). Once modeled in DICOM format, we compared the percent difference between the heights of age-scaled DICOM phantoms and heights (50 th percentile) reported by the United States Centers for Disease Control and Prevention (CDC) for male and female children of the same ages. To validate the conversion of the phantom from FORTRAN to DICOM format across the full age range, we compared the percent difference between the volume of body regions (e.g. head, neck, and trunk) for 0.1, 0.5, 1, 2, 3, 5, 8, 10, 15 and 18 years for phantoms coded in the two formats. Lastly, we calculated the normalized mean square distance (NMSD) between the organs (heart, liver, stomach, lungs, and brain) of both phantom formats for the aforementioned ages. Results The heights of our age-scaled phantom and CDC reported heights agreed within 7% from infant to adult for both genders, with agreement better than 2% for ages five and older (Fig 1D/Table 1). The percent difference between the volume of head, neck and trunk of our phantoms in FORTRAN and DICOM formats were in good agreement, within. For all of the ages, the NMSDs were 0.0mm for each organ except brain where NMSDs are non-zero but less than 0.5mm agreement (Table 1).

Conclusion We successfully developed the 3D model of the phantom in DICOM format and validated it with our previous model. The phantom in DICOM format can be imported into any commercial treatment planning system for whole body dosimetry studies. PO-1331 Monte Carlo assessment of the PTW-31021 Semiflex 3D performance under a 0.35 tesla magnetic field G.V. SanturiO 1 , S. Blak Nyrup Biancardo 1 , U. Bjelkengren 1 1 Herlev Hospital, Radiotherapy, Herlev, Denmark Purpose or Objective Magnetic resonance (MR) imaging systems in radiotherapy offer the possibility of acquiring high quality images with high soft-tissue contrast without adding extra dose to the patient. The MR system combined with a linear accelerator allows for on-table adaptive radiotherapy and also gated treatment using live images as the patient is being treated. This in turn may allow for margin reductions and dose escalation. However, the implementation increases the challenges for the medical physicist as the dosimetry under magnetic fields has not been incorporated in the currently used radiotherapy protocols (TRS-398, TG-51, TRS-483). The objective of this study is to investigate the performance of the ionization chamber PTW-31021 Semiflex 3D under a 0.35 tesla magnetic field for reference dosimetry using the magnetic field correction factor. Material and Methods The magnetic field correction factor ( k B ) was calculated using Monte Carlo simulations. The Monte Carlo based toolkit EGSnrc was used for the computations. The radiation source used was a 6 MV flattening filter free beam as this study aims to incorporate experimental measurements using a ViewRay MRIdian accelerator. The impact of the magnetic field was evaluated with the