CALCULATION OF X-RAY-SPECTRA FOR RADIOSURGICAL BEAMS

Citation
Ke. Sixel et Ba. Faddegon, CALCULATION OF X-RAY-SPECTRA FOR RADIOSURGICAL BEAMS, Medical physics, 22(10), 1995, pp. 1657-1661
Citations number
20
Categorie Soggetti
Radiology,Nuclear Medicine & Medical Imaging
Journal title
ISSN journal
00942405
Volume
22
Issue
10
Year of publication
1995
Pages
1657 - 1661
Database
ISI
SICI code
0094-2405(1995)22:10<1657:COXFRB>2.0.ZU;2-9
Abstract
As conformal radiosurgery using micromultileaf collimators gains feasi bility, dose calculation algorithms based on Monte Carlo or convolutio n techniques may become necessary. These require radiosurgical x-ray s pectra. The most accurate method currently available to estimate clini cal radiosurgery spectra is the Monte Carlo method. In this study the EGS4 Monte Carlo system was used to simulate the thick target of a 6 M V linear accelerator used for radiosurgery in our center. The calculat ed spectrum was attenuated through any significant mass thickness of m aterial downstream from the target. The attenuated thick-target spectr al distributions calculated both with and without the flattening filte r were compared to the attenuated, thin target spectrum based on the s mall angle Schiff analytical spectrum calculated for the same target a nd attenuator material, as well as with a published spectrum from a fu ll Monte Carlo simulation of a treatment head with a flattener in plac e. The Schiff spectrum neglects contributions from lower-energy scatte red electrons that significantly degrade the quality of the beam. The flattener is removed from our accelerator during radiosurgery to incre ase the dose rate to approximately 750 cGy/min for a 10X10 cm(2) field at the depth of dose maximum. This leaves a substantial fluence of ph otons below 1 MeV that are not observed in published spectra calculate d for accelerators with flattening filters. Removal of the flattening filter has a measurable effect on the central axis depth dose, reducin g the percentage dose at 10 cm depth from 59.2% to 54.3% for a 10 mm d iam field. Radiosurgical off-axis ratios and percentage depth dose dis tributions calculated from these spectra with the EGS4 Monte Carlo cod e were compared to measured data. Measured and calculated dose distrib utions both with and without flattener were in good agreement. The dos e distributions were found to be insensitive to the differences in the various calculated spectral distributions. Thus, although the attenua ted Schiff spectrum is significantly harder than the clinical beam, it is adequate for dose calculations of radiosurgical beams.