ESTRO 38 Abstract book

S976 ESTRO 38

patients with different tumor locations, applicator sizes, and with different energies. Results MC model validation process ended with good match between measurement and MC PDDs, Profiles and Output factors as the maximum value of the root mean square deviation (RMSD) was less than 3% between them. Absolut and relative doses in PTVs and in Organ at risks were verified using DVH and color wash like any classical linear accelerator TPS. Alhamada et al (2018) and Kamomae et al (2017) have studied new shielding disks and tungsten paper for organs at risk protection, but they did their studies in water phantom. It is possible to study any shielding disk design in clinical usage. Dosimetry impact verification of using customized applicator or beam shaper is available using our methodology. Summation of 3D IOERT dose with other radiotherapy plans is possible and helpful for total overdose avoidance. Conclusion This study provides a step forward for dosimetry verification of IOERT patients. More accurate dosimetry could be achieved by using DICOM images of CBCT inside the operation room which it will be available next year in our clinic. EP-1803 Advances in the Patient Specific QA applied to VMAT and Tomotherapy M. Picioli 1 , K. Torzsok 1 , A. Ruiz Plata 1 , F. Marangoni 1 , J. Aponte 1 , H. Broque 1 1 Fundacion Arturo Lopez Perez, Medical Physics, Santiago, Chile Purpose or Objective It has been more than 20 years (1998) that patient specific quality assurance (PSQA) is playing an important role to verify plan deliverability and reliability for modulated treatments. Since then, there have been massive improvements in machine accuracy in terms of mechanic and dosimetric parameters. Likewise TPS algorithms and experience with these techniques has grown exponentially. However, the verification methodologies kept being almost the same over these years. In this work we have shown the factibility of an alternative prediction index for potencial plan failure for Tomotherapy and VMAT. Material and Methods A restropective analysis over 1000 patients already treated was carried out for an Elekta Synergy equiped with Agiliy MLC and a Tomotherapy HD. Patient Specific QA have been measured for every single patient before treatment and results from both machines were analyzed and registered using IBA MatriXX Evolution and the criteria of 3%, 3mm, 10% threshold have been applied. Although both machines have different delivery procedures, the modulation factor calculated during planning have being shown to be the main predictive factor of plan complexity and had a direct impact on gamma results. As an example, the following equation describes the way the Modulation factor is calculated for VMAT plans:

As expected, our results show that with decreasing detector size, the measured penumbra width and therefore the distance to the point of inflection decreases [2]. Considering all energies, the mean time duration for linac output constancy daily measurement using EPID and IC was respectively 16 and 30 minutes. The mean time duration for beam parameters measurement using EPID and IC was respectively 14 and 200 minutes.

Conclusion The use of automatic EPID based beam QA dramatically decreases the time dedicated to QA, eliminates manual steps in the process, reducing human errors, improving efficiency, robustness and productivity for radiotherapy providers with a significant decrease in waiting time for patients. In our institution we have chosen to use daily EPID QA to verify output constancy and beam uniformity without any modification on quality or tolerances of our QA process. [1] Boutry C, Sors A, Fontaine J, Delaby N, Delpon G. A simple algorithm to convert EPID gray values into absorbed dose to water without prior knowledge. Med Phys. 2017 Dec ; 44(12):6647-6653. [2] Laub WU and WongT The volume effect of detectors in the dosimetry of small fields used in IMRT Med Phys. 2003 Mar ; 30(3):341-347 EP-1802 Dosimetry verification of IntraOperative Radiation Therapy (IORT): a Monte Carlo Study H. Alhamada 1 , S. Simon 2 , C. Philippson 2 , C. Vandekerkhove 2 , Y. Jourani 2 , N. Pauly 3 , N. Reynaert 2 , D. Van Gestel 2 1 Institut Jules Bordet ULB, Radiotherapy, Brussels, Belgium ; 2 Institute Jules Bordet, Radiotherapy Department, Brussels, Belgium ; 3 Nuclear Metrology, Université Libre de Bruxelles, Brussels, Belgium Purpose or Objective Most of intraOperative Radiation Therapy (IORT) clinics have not a treatment planning system (TPS). High dose (up to 24 Gy) is given in one session depending only on manual calculation. This study investigates the feasibility of doing 3D dosimetry of IORT patients using post operation patient’s CT images with Monte Carlo (MC) code. Material and Methods Mobetron 1000® is a mobile linear accelerator was modeled using BEAMnrc MC code. MC model validation was done for 12, 9 and 6 MeV energies by comparing MC results with measurements according to MC model tuning of Laccarino et al (2011). Absolute dose calibration was done using methodology of Popescu et al (2005).CTCREATE code and home-made code were used to convert post operation CT DICOM images to MC phantom. Each voxel in patient’s MC phantom is one of these materials (Air521ICRU, Lung521ICRU, Tissue521ICRU, Bone521ICRU, PMMA521ICRU, AL521ICRU, Steel521ICRU). Contouring of the planning target volume (PTV) and the organs at risk was done by the radiation oncologist. IORT parameters (applicator size and position, energy, Monitors Units, shielding disk, Bolus, ….) as they were used in the operation room simulated using BEAMnrc and DOSXYZnrc for each patient. Finally, dose calculation was done in each patient’s MC phantom. This study was done for 20

On the other hand, for Tomotherapy, leaf open time (LOT) is the major factor used for calculating modulation in the TPS. Regardless the system being used, reducing the modulation during planning might be the most effective way of having more representative plans with better gamma values. For a better understanding on how it would affect QA results, modulation factors were classified in