Monte Carlo calculated beam quality correction factors for high energy electron beams.

PURPOSE: As the dosimetry protocol TRS 398 is being revised and the ICRU report 90 provides new recommendations for density correction as well as the mean ionization energies of water and graphite, updated beam quality correction factors kQ are calculated for reference dosimetry in electron beams and for independent validation of previously determined values. METHODS: Monte Carlo simulations have been performed using EGSnrc to calculate the absorbed dose to water and the dose to the active volumes of ionization chambers SNC600c, SNC125c and SNC350p (all Sun Nuclear, A Mirion Medical Company, Melbourne, FL). Realistic clinical electron beam spectra were used to cover the entire energy range of therapeutic electron accelerators. The Monte Carlo simulations were validated by measurements on a clinical linear accelerator. With regards to the cylindrical chambers, the simulations were performed according to the setup recommendations of TRS 398 and AAPM TG 51, i.e. with and without consideration of a reference point shift by rcav/2. RESULTS: kQ values as a function of the respective beam quality specifier R50 were fitted by recommended equations for electron beam dosimetry in the range of 5 MeV to 18 MeV. The fitting curves to the calculated values showed a root mean square deviation between 0.0016 and 0.0024. CONCLUSION: Electron beam quality correction factors kQ were calculated by Monte Carlo simulations for the cylindrical ionization chambers SNC600c and SNC125c as well as the plane parallel ionization chamber SNC350p to provide updated data for the TRS 398 and TG 51 dosimetry protocols.

Validation of an EGSnrc-based Monte Carlo model for a complex 2D-array for technical QA measurements of a linear accelerator.

PURPOSE: Multi-axis ionization chamber arrays can be used for quality assurance (QA) and measurement of linear accelerator (linac) specific data. In this work, the ability of the IC Profiler (Sun Nuclear Corp., Melbourne, Florida) detector array to measure the photon beam quality specifier %dd(10) x $_\textrm {{\it x}}$ and TPR20, 10 was investigated. To investigate the method for beam energy QA using a two-dimensional detector array, a Monte Carlo-based model of the detector array was developed and validated. METHODS: A Monte Carlo-based model of the IC Profiler detector array with Quad Wedge accessories was developed in detail from drawings provided by the manufacturer using the egs++ class library from the EGSnrc code system. Monte Carlo simulations were used to calculate the absorbed dose in the 251 ionization chambers of the IC profiler in the 6 MV Elekta Precise radiation field. To validate the results from the Monte Caro simulations, measurements were performed on clinical 6 MV linacs. To vary the photon beam quality of the Elekta 6 MV linac, the current of the bending magnet was varied. Furthermore, the area ratio A R $AR$ was calculated from IC Profiler measurements with the Quad Wedge accessories. RESULTS: Measurements as well as Monte Carlo simulations confirmed the linear relationship between the area ratio A R $AR$ and the investigated photon beam quality specifier %dd(10) x $_\textrm {{\it x}}$ and TPR20, 10 for the investigated radiation source. Furthermore, the Monte Carlo-simulated data were within the 95% confidence interval of the linear fit to the measured data. This enabled the Monte Carlo-based IC Profiler model to be used for further investigations. The A R $AR$ values were calculated for various electron beam sizes and the angle of incidence on the target of the linac. CONCLUSIONS: A Monte-Carlo-based model of the detector array was developed, which could successfully reproduce measurements, demonstrating that even very complex geometries can be modeled in EGSnrc. Moreover, the study showed that the validated Monte Carlo model has the potential to investigate variations in beam parameters and their effects on AR ratios and %dd(10) x $_\textrm {{\it x}}$ that may not be investigated experimentally. While these findings may help users gain a deeper understanding of the QA method, the Monte Carlo model enables other complex investigations, such as the simulation of measurements in the presence of magnetic fields, or the simulation of measurements on novel treatment delivery techniques and devices.

Experimental and Monte Carlo-based determination of magnetic field correction factors k B , Q $k_{B,Q}$ in high-energy photon fields for two ionization chambers.

BACKGROUND: The integration of magnetic resonance tomography into clinical linear accelerators provides high-contrast, real-time imaging during treatment and facilitates online-adaptive workflows in radiation therapy treatments. The associated magnetic field also bends the trajectories of charged particles via the Lorentz force, which may alter the dose distribution in a patient or a phantom and affects the dose response of dosimetry detectors. PURPOSE: To perform an experimental and Monte Carlo-based determination of correction factors k B , Q $k_{B,Q}$ , which correct the response of ion chambers in the presence of external magnetic fields in high-energy photon fields. METHODS: The response variation of two different types of ion chambers (Sun Nuclear SNC125c and SNC600c) in strong external magnetic fields was investigated experimentally and by Monte Carlo simulations. The experimental data were acquired at the German National Metrology Institute, PTB, using a clinical linear accelerator with a nominal photon energy of 6 MV and an external electromagnet capable of generating magnetic flux densities of up to 1.5 T in opposite directions. The Monte Carlo simulation geometries corresponded to the experimental setup and additionally to the reference conditions of IAEA TRS-398. For the latter, the Monte Carlo simulations were performed with two different photon spectra: the 6 MV spectrum of the linear accelerator used for the experimental data acquisition and a 7 MV spectrum of a commercial MRI-linear accelerator. In each simulation geometry, three different orientations of the external magnetic field, the beam direction and the chamber orientation were investigated. RESULTS: Good agreement was achieved between Monte Carlo simulations and measurements with the SNC125c and SNC600c ionization chambers, with a mean deviation of 0.3% and 0.6%, respectively. The magnitude of the correction factor k B , Q $k_{B,Q}$ strongly depends on the chamber volume and on the orientation of the chamber axis relative to the external magnetic field and the beam directions. It is greater for the SNC600c chamber with a volume of 0.6 cm3 than for the SNC125c chamber with a volume of 0.1 cm3 . When the magnetic field direction and the chamber axis coincide, and they are perpendicular to the beam direction, the ion chambers exhibit a calculated overresponse of less than 0.7(6)% (SNC600c) and 0.3(4)% (SNC125c) at 1.5 T and less than 0.3(0)% (SNC600c) and 0.1(3)% (SNC125c) for 0.35 T for nominal beam energies of 6 MV and 7 MV. This chamber orientation should be preferred, as k B , Q $k_{B,Q}$ may increase significantly in other chamber orientations. Due to the special geometry of the guard ring, no dead-volume effects have been observed in any orientation studied. The results show an intra-type variation of 0.17% and 0.07% standard uncertainty (k=1) for the SNC125c and SNC600c, respectively. CONCLUSION: Magnetic field correction factors k B , Q $k_{B,Q}$ for two different ion chambers and for typical clinical photon beam qualities were presented and compared with the few data existing in the literature. The correction factors may be applied in clinical reference dosimetry for existing MRI-linear accelerators.

Monte Carlo calculated beam quality correction factors for two cylindrical ionization chambers in photon beams.

PURPOSE: Although several studies provide data for reference dosimetry, the SNC600c and SNC125c ionization chambers (Sun Nuclear Corporation, Melbourne, FL) are in clinical use worldwide for which no beam quality correction factors kQ are available. The goal of this study was to calculate beam quality correction factors kQ for these ionization chambers according to dosimetry protocols TG-51, TRS 398 and DIN 6800-2. METHODS: Monte Carlo simulations using EGSnrc have been performed to calculate the absorbed dose to water and the dose to air within the active volume of ionization chamber models. Both spectra and simulations of beam transport through linear accelerator head models were used as radiation sources for the Monte Carlo calculations. RESULTS: kQ values as a function of the respective beam quality specifier Q were fitted against recommended equations for photon beam dosimetry in the range of 4 MV to 25 MV. The fitting curves through the calculated values showed a root mean square deviation between 0.0010 and 0.0017. CONCLUSIONS: The investigated ionization chamber models (SNC600c, SNC125c) are not included in above mentioned dosimetry protocols, but are in clinical use worldwide. This study covered this knowledge gap and compared the calculated results with published kQ values for similar ionization chambers. Agreements with published data were observed in the 95% confidence interval, confirming the use of data for similar ionization chambers, when there are no kQ values available for a given ionization chamber.

Reference conditions for ion-chamber based HDR brachytherapy dosimetry and for the calibration of high-resolution solid detectors.

The aim of this study has been to develop a two-step method of in-phantom dosimetry around a brachytherapy 192Ir photon source. The first step is to measure the absorbed dose rate to water with a calibrated ionization chamber under reference conditions, the second to cross-calibrate, under these conditions, small solid-state detectors such as silicon diodes, synthetic diamond or scintillation detectors suited for spatially resolved dose rate measurements at other, particularly at smaller source axis distances in the water phantom. This two-step approach constitutes a method for in-phantom dosimetry in brachytherapy, analogous to the "small calibration field" commonly used in teletherapy to provide the reference conditions for the cross-calibration of high-resolution detectors. Under reference conditions, all known corrections for radiation quality, volume averaging and position of the chamber's effective point of measurement (EPOM) have to be applied. The study is therefore particularly devoted to (1) the experimental determination of the position of the source axis, (2) a general formulation for the volume averaging correction factor of small ionization chambers and (3) the experimental determination of the EPOM positions for the PinPoint chamber 31014 and the 3D-PinPoint chamber PTW 31022 (both PTW Freiburg, Germany). The distance of 30mm from the source axis was chosen as the reference condition for cross calibrations. This concept is realized with the instrumentation available in a hospital, a scanning-type water phantom, a software package for small field dosimetry and detectors typically used in clinical routine dosimetry. The present development of a method of in-phantom dose measurement under 192Ir brachytherapy conditions was performed in recognition of the primary role of dose calculations, e.g. according to the AAPM TG43 recommendations. But in addition, the methodology tested here is paving a practicable way for the experimental check of typical dose values under clinical conditions, should the need arise.

Magnetic fields are causing small, but significant changes of the radiochromic EBT3 film response to 6 MV photons.

The optical density (OD) of EBT3 radiochromic films (Ashland Specialty Ingredients, Bridgewater, NJ, USA) exposed to absorbed doses to water up to D = 20 Gy in magnetic fields of B = 0.35 and 1.42 T was measured in the three colour channels of an Epson Expression 10000XL flatbed scanner. A 7 cm wide water phantom with fixed film holder was placed between the pole shoes of a constant-current electromagnet with variable field strength and was irradiated by a 6 MV photon beam whose axis was directed at right angles with the field lines. The doses at the film position at water depth 5 cm were measured with a calibrated ionization chamber when the magnet was switched off and were converted to the doses in presence of the magnetic field via the monitor units and by a Monte Carlo-calculated correction accounting for the slight change of the depth dose curves in magnetic fields. In the presence of the 0.35 and 1.42 T fields small negative changes of the OD values at given absorbed doses to water occurred and just significantly exceeded the uncertainty margin given by the stochastic and the uncorrected systematic deviations. This change can be described by a +2.1% change of the dose values needed to produce a given optical density in the presence of a 1.42 T field. The thereby modified OD versus D function remained unchanged irrespective of whether the original short film side-the preference direction of the monomer crystals of the film-was directed parallel or orthogonal to the magnetic field. The 'orientation effect', the difference between the optical densities measured in the 'portrait' or 'landscape' film positions on the scanner bed caused by the reflection of polarised light in the scanner's mirror system, remained unaltered after EBT3 film exposure in magnetic fields. An independent optical bench investigation of EBT3 films exposed to doses of 10 and 20 Gy at 0.35 and 1.42 T showed that the direction of the electric vector of polarised light experiencing the largest transmission through EBT3 films remained unaltered after film exposure in the magnetic fields. The observed small modification of the OD versus D curve of the radiochromic film EBT3 in the range up to 20 Gy and 1.42 T, hardly exceeding the experimental uncertainty margin, numerically confirms other recent studies on EBT3 film. A stronger magnetic field effect had been observed with the previous product EBT2 exposed to 60Co gamma radiation at 0.35 T.

Evaluation of water-mimicking solid phantom materials for use in HDR and LDR brachytherapy dosimetry.

In modern HDR or LDR brachytherapy with photon emitters, fast checks of the dose profiles generated in water or a water-equivalent phantom have to be available in the interest of patient safety. However, the commercially available brachytherapy photon sources cover a wide range of photon emission spectra, and the range of the in-phantom photon spectrum is further widened by Compton scattering, so that the achievement of water-mimicking properties of such phantoms involves high requirements on their atomic composition. In order to classify the degree of water equivalence of the numerous commercially available solid water-mimicking phantom materials and the energy ranges of their applicability, the radial profiles of the absorbed dose to water, D w, have been calculated using Monte Carlo simulations in these materials and in water phantoms of the same dimensions. This study includes the HDR therapy sources Nucletron Flexisource Co-60 HDR (60Co), Eckert und Ziegler BEBIG GmbH CSM-11 (137Cs), Implant Sciences Corporation HDR Yb-169 Source 4140 (169Yb) as well as the LDR therapy sources IsoRay Inc. Proxcelan CS-1 (131Cs), IsoAid Advantage I-125 IAI-125A (125I), and IsoAid Advantage Pd-103 IAPd-103A (103Pd). Thereby our previous comparison between phantom materials and water surrounding a Varian GammaMed Plus HDR therapy 192Ir source (Schoenfeld et al 2015) has been complemented. Simulations were performed in cylindrical phantoms consisting of either water or the materials RW1, RW3, Solid Water, HE Solid Water, Virtual Water, Plastic Water DT, Plastic Water LR, Original Plastic Water (2015), Plastic Water (1995), Blue Water, polyethylene, polystyrene and PMMA. While for 192Ir, 137Cs and 60Co most phantom materials can be regarded as water equivalent, for 169Yb the materials Plastic Water LR, Plastic Water DT and RW1 appear as water equivalent. For the low-energy sources 106Pd, 131Cs and 125I, only Plastic Water LR can be classified as water equivalent.

Changes of the optical characteristics of radiochromic films in the transition from EBT3 to EBT-XD films.

A new type of radiochromic film, the EBT-XD film, has been introduced with the aim to reduce the orientation effect and the lateral response artifact occurring in the use of radiochromic films together with flatbed scanners. The task of the present study is to quantify the changes of optical characteristics involved with the transition from the well-known EBT3 films to the new EBT-XD films, using the optical bench arrangement already applied by Schoenfeld et al (2014 Phys. Med. Biol. 59 3575-97). Largely reduced polarization effects and the almost complete loss of the anisotropy of the scattered light produced in a radiation-exposed film have been observed. The Rayleigh-Debye-Gans theory is used to understand these optical changes as arising from the reduced length-to-width ratio of the LiPCDA polymer crystals in the active layer of the EBT-XD film. The effect of these changes on the flatbed scanning artifacts will be shortly addressed, but treated in more detail in a further paper.

The origin of the flatbed scanner artifacts in radiochromic film dosimetry-key experiments and theoretical descriptions.

The optical origin of the lateral response and orientation artifacts, which occur when using EBT3 and EBT-XD radiochromic films together with flatbed scanners, has been reinvestigated by experimental and theoretical means. The common feature of these artifacts is the well-known parabolic increase in the optical density OD(x) = -log10 I(x)/I 0(x) versus offset x from the scanner midline (Poppinga et al 2014 Med. Phys. 41 021707). This holds for landscape and portrait orientations as well as for the three color channels. Dose-independent optical subjects, such as neutral density filters, linear polarizers, the EBT polyester foil and diffusive glass, also present the parabolic lateral artifact when scanned with a flatbed scanner. The curvature parameter c of the parabola function OD(x) = c 0 + cx 2 is found to be a linear function of the dose, the parameters of which are influenced by the film orientation and film type, EBT3 or EBT-XD. The ubiquitous parabolic shape of function OD(x) is attributed (a) to the optical path-length effect (van Battum et al 2016 Phys. Med. Biol. 61 625-49), due to the increasing obliquity of the optical scanner light associated with increasing offset x from the scanner midline, and (b) and (c) to the partial polarization and scattering of the light leaving the film, which affect the ratio [Formula: see text], thus making OD(x) increase with x 2. The orientation effect results from the changes of effects (b) and (c) associated with turning the film position, and thereby the orientation of the polymer structure of the sensitive film layer. In a comparison of experimental results obtained with selected optical subjects, the relative weights of the contributions of the optical path-length effect and the polarization and scattering of light leaving the films to the lateral response artifact have been estimated to be of the same order of magnitude. Mathematical models of these causes for the parabolic shape of function OD(x) are given as appendices.

Water equivalent phantom materials for ¹9²Ir brachytherapy.

Several solid phantom materials have been tested regarding their suitability as water substitutes for dosimetric measurements in brachytherapy with (192)Ir as a typical high energy photon emitter. The radial variations of the spectral photon fluence, of the total, primary and scattered photon fluence and of the absorbed dose to water in the transversal plane of the tested cylindrical phantoms surrounding a centric and coaxially arranged Varian GammaMed afterloading (192)Ir brachytherapy source were Monte-Carlo simulated in EGSnrc. The degree of water equivalence of a phantom material was evaluated by comparing the radial dose-to-water profile in the phantom material with that in water. The phantom size was varied over a large range since it influences the dose contribution by scattered photons with energies diminished by single and multiple Compton scattering. Phantom axis distances up to 10 cm were considered as clinically relevant. Scattered photons with energies reaching down into the 25 keV region dominate the photon fluence at source distances exceeding 3.5 cm. The tested phantom materials showed significant differences in the degree of water equivalence. In phantoms with radii up to 10 cm, RW1, RW3, Solid Water, HE Solid Water, Virtual Water, Plastic Water DT, and Plastic Water LR phantoms show excellent water equivalence with dose deviations from a water phantom not exceeding 0.8%, while Original Plastic Water (as of 2015), Plastic Water (1995), Blue Water, polyethylene, and polystyrene show deviations up to 2.6%. For larger phantom radii up to 30 cm, the deviations for RW1, RW3, Solid Water, HE Solid Water, Virtual Water, Plastic Water DT, and Plastic Water LR remain below 1.4%, while Original Plastic Water (as of 2015), Plastic Water (1995), Blue Water, polyethylene, and polystyrene produce deviations up to 8.1%. PMMA plays a separate role, with deviations up to 4.3% for radii not exceeding 10 cm, but below 1% for radii up to 30 cm. As suggested by the results of the dose simulations and the values of the linear attenuation coefficient, µ, over a large energy range, the balanced content of inorganic additives in a phantom material is regarded as the key feature, providing water equivalence with regard to the attenuation of the primary photons, the release of low-energy photons by Compton scattering, and their attenuation by a combination of the photoelectric and Compton effects.

The artefacts of radiochromic film dosimetry with flatbed scanners and their causation by light scattering from radiation-induced polymers.

Optical experiments and theoretical considerations have been undertaken in order to understand the causes of the 'orientation effect' and the 'parabola effect', the artefacts impairing the desired light absorption measurement on radiochromic EBT3 films with flatbed scanners. EBT3 films exposed to doses up to 20.9 Gy were scanned with an Epson Expression 10000XL flatbed scanner in landscape and portrait orientation. The horizontally and vertically polarized light components of the scanner were determined, and another Epson Expression 10000XL flatbed scanner was disassembled to examine its optical components. The optical properties of exposed and unexposed EBT3 films were studied with incident polarized and unpolarized white light, and the transmitted red light was investigated for its polarization and scattering properties including the distribution of the scattering angles. Neutral density filters were studied for comparison. Guidance was sought from the theory of light scattering from rod-like macromolecular structures. The drastic dose-dependent variation of the transmitted total light current as function of the orientation of front and rear polarizers, interpreted by light scattering theory, shows that the radiation-induced polymerization of the monomers of EBT3 films produces light scattering oscillators preferably polarized at right angles with the coating direction of the film. The directional distribution of the scattered light is partly anisotropic, with a preferred scattering plane at right angles with the coating direction, indicating light scattering from stacks of coherently vibrating oscillators piled up along the monomer crystals. The polyester carrier film also participates in these effects. The 'orientation' and 'parabola' artefacts due to flatbed scanning of radiochromic films can be explained by the interaction of the polarization-dependent and anisotropic light scattering from exposed and unexposed EBT3 films with the quantitative difference between the scanner's horizontally and vertically polarized light supply and with the limited directional acceptance of the scanner's light recording system.