Monte Carlo simulation of beam characteristics from small fields based on TrueBeam flattening-filter-free mode.

Zhongsu Feng,Yibao Zhang,Haizhen Yue,Jinsheng Cheng,Hao Wu,Xu Su

doi:10.1186/s13014-016-0601-2

Abstract

PurposeThrough the Monte Carlo (MC) simulation of 6 and 10 MV flattening-filter-free (FFF) beams from Varian TrueBeam accelerator, this study aims to find the best incident electron distribution for further studying the small field characteristics of these beams.MethodsBy incorporating the training materials of Varian on the geometry and material parameters of TrueBeam Linac head, the 6 and 10 MV FFF beams were modelled using the BEAMnrc and DOSXYZnrc codes, where the percentage depth doses (PDDs) and the off-axis ratios (OARs) curves of fields ranging from 4 × 4 to 40 × 40 cm2 were simulated for both energies by adjusting the incident beam energy, radial intensity distribution and angular spread, respectively. The beam quality and relative output factor (ROF) were calculated. The simulations and measurements were compared using Gamma analysis method provided by Verisoft program (PTW, Freiburg, Germany), based on which the optimal MC model input parameters were selected and were further used to investigate the beam characteristics of small fields.ResultsThe Full Width Half Maximum (FWHM), mono-energetic energy and angular spread of the resultant incident Gaussian radial intensity electron distribution were 0.75 mm, 6.1 MeV and 0.9° for the nominal 6 MV FFF beam, and 0.7 mm, 10.8 MeV and 0.3° for the nominal 10 MV FFF beam respectively. The simulation was mostly comparable to the measurement. Gamma criteria of 1 mm/1 % (local dose) can be met by all PDDs of fields larger than 1 × 1 cm2, and by all OARs of no larger than 20 × 20 cm2, otherwise criteria of 1 mm/2 % can be fulfilled. Our MC simulated ROFs agreed well with the measured ROFs of various field sizes (the discrepancies were less than 1 %), except for the 1 × 1 cm2 field.ConclusionsThe MC simulation agrees well with the measurement and the proposed model parameters can be clinically used for further dosimetric studies of 6 and 10 MV FFF beams.

Highlights

To facilitate the model development and dose computation, conventional radiation beam was flattened through the filter mounted in the gantry head
Belosi et al [11] evaluated the accuracy of the distributed phase-space files for flattening filter free (FFF) beams by comparing them with experimental measurements based on ten TrueBeam systems, and concluded that the phase-space files can be used for accurate Monte Carlo (MC) dose estimation, their applications to MC simulation were limited
The photon and electron cut off energy (PCUT and ECUT) values were set to 0.01 and 0.521 MeV, respectively

Summary

Introduction

To facilitate the model development and dose computation, conventional radiation beam was flattened through the filter mounted in the gantry head. Accurate geometric and material parameters of the Linac head are critical for the MC modelling, yet they have not been made available for TrueBeam except for the first and second generation phase-space files. Belosi et al [11] evaluated the accuracy of the distributed phase-space files for FFF beams by comparing them with experimental measurements based on ten TrueBeam systems, and concluded that the phase-space files can be used for accurate MC dose estimation, their applications to MC simulation were limited. Relative to the Varian phase-space files, the ansatz geometry reproduced the measured dose more accurately, but the thin filters were made of high Z materials that increased the head scatter and affected the beam quality

Objectives

Methods

Results

Conclusion