Encrypted login | home

Program Information

Generation of 3D Dosimetric Reference Datasets for Commissioning and Validation of 192Ir Brachytherapy Model-Based Dose Calculation Software

Y Ma

Y Ma1*, J Vijande2 , F Ballester3 , Asa Carlsson Tedgren4 , D Granero5 , A Haworth6 , F Mourtada7 , G Fonseca8 , K Zourari9 , P Papagiannis10 , M Rivard11 , F Siebert12 , R Sloboda13 , R Smith14 , R Thomson15 , F Verhaegen16 , L Beaulieu17 , (1) CHU de Quebec, Quebec, QC, (2) University of Valencia, Burjassot, ,(3) University of Valencia, Burjassot, ,(4) Linkoping Univ, Linkoping, ,(5) ERESA-Hospital General Universitario, Mislata, ,(6) Peter MacCallum Cancer Centre, Vic Australia, ,(7) Christiana Care Hospital, Newark, DE, (8) Instituto de Pesquisas Energeticas e Nucleares IPEN-CNEN/SP, Sao Paulo, Maastricht, (9) Medical Physics Laboratory, Medical School, University of Athens, Athens, Athens, (10) Univ Athens, Athens, ,(11) ,Tufts Univ. School Med., Boston, MA, (12) Clinic of Radiotherapy (Radiooncology), Kiel, ,(13) Cross Cancer Institute, Edmonton, AB, (14) St. Thomas Hospital, Nashville, TN, (15) Carleton Univ, Ottawa, ON, (16) Maastro Clinic, Maastricht, ,(17) Centre Hospitalier Univ de Quebec, Quebec, QC


TH-AB-BRA-2 (Thursday, July 16, 2015) 7:30 AM - 9:30 AM Room: Ballroom A

Previously, generic HDR ¹⁹²Ir source and shielded applicator were designed by WG-MBDCA, the AAPM-ESTRO-ABG Working Group on model-based dose calculation algorithms. TPS vendors have incorporated both into their TPS. A charge of WG-MBDCA is to develop a limited number of well-defined test cases and perform MBDCA dose calculations and comparisons. Here, we describe the test cases proposed, the 3D dosimetric data generation process and the 3D gamma analysis performed.

A voxelized phantom of 512x512x512 voxels (HU=-1024,Air), with each voxel being 1x1x1 mm³, containing a concentric cubic contour structure (HU=0,Water) of dimension 201x201x201 voxels, was created in the DICOM-RT format. Three configurations were tested: i) source located at the center with surrounding structure assigned to water to mimic TG-43 conditions, ii) source displaced +7 cm along the x axis, iii) source located at the center with the shielded applicator in place. Advanced Collapsed-cone Engine (ACE, Elekta) and AcurosBV (Varian) were benchmarked. Reference datasets were independently produced by different groups using BrachyDose (10¹⁰ photons), GEANT4(4x10¹⁰), Penelope (2x10¹⁰), MCNP5(2 x10¹⁰), MCNP6(10¹¹), and ALGEBRA(4x10¹⁰) Monte Carlo (MC) dose calculation platforms.

For the three configurations, MC agreed within 1% for the volume bounded by the 1% isodose surface; MCNP6 data sets were designated as reference (statistical error < 0.2% across the phantom). A 2%/1mm 3D γ-index criteria was used, within the volume bounded by 5% isodose surface, to account for both dose difference and source placement errors in TPS. Voxel pass rates were 95.2%, 99.5% and 90% for ACE and 98%, 98.4% and 95.9% for AcurosBV, respectively.

The evaluated test cases are suitable for MBDCA QA. DICOM-RT files are being integrated into the joint AAPM/IROC Houston registry to support end-user TPS software commissioning.

Contact Email: