EGS4 Code System Research Articles

WE‐C‐AUD‐03: Investigation of Fast Monte Carlo Dose Calculation for CyberKnife SRS/SRT Treatment Planning

Purpose: Advanced stereotactic radiosurgery (SRS) and stereotactic radiotherapy (SRT) treatments require accurate dose calculation for treatment planning especially for treatment sites involving heterogeneous patient anatomy. In this work, we have implemented a fast Monte Carlo dose calculation algorithm for SRS/SRT treatment planning with the CyberKnife® system. Methods and Materials: Our system employs a superposition Monte Carlo algorithm. Photon mean free paths and interaction types for different materials and energies as well as the tracks of secondary electrons are pre‐simulated using the EGS4 code system. Photon interaction forcing and splitting are applied to the source photons in a patient calculation and the pre‐simulated tracks are repeated with proper corrections based on the tissue density and electron stopping powers. Electron energy is deposited along the tracks and accumulated in every voxel of the simulation geometry. Scattered and bremsstrahlung photons are transported, after applying the Russian Roulette technique, in the same way as the primary photons. Dose calculations are compared with full Monte Carlo simulations and the CyberKnife treatment planning system (TPS) for lung and head & neck treatments. Results: Comparisons with full Monte Carlo simulations show excellent agreement (within 0.5%). Significant differences in the target dose are found between Monte Carlo simulations and the CyberKnife TPS for SRS lung treatment. The calculation time using our superposition Monte Carlo algorithm is reduced up to 62 times (46 times on average for 10 typical clinical cases) compared to full Monte Carlo simulations. Conclusions: SRS/SRT dose distributions calculated by simple dose algorithms may be significantly overestimated for small lung target volumes, which can be improved by accurate Monte Carlo dose calculations. Properly implemented fast Monte Carlo algorithms can improve dosimetric accuracy with little or no compromise to computational efficiency.

Film dosimetry calibration method for pulsed‐dose‐rate brachytherapy with an source

192Ir sources have been widely used in clinical brachytherapy. An important challenge is to perform dosimetric measurements close to the source despite the steep dose gradient. The common, inexpensive silver halide film is a classic two-dimensional integrator dosimeter and would be an attractive solution for these dose measurements. The main disadvantage of film dosimetry is the film response to the low-energy photon. Since the photon energy spectrum is known to vary with depth, the sensitometric curves are expected to be dependent on depth. The purpose of this study is to suggest a correction method for silver halide film dosimetry that overcomes the response changes at different depths. Sensitometric curves have been obtained at different depths with verification film near a 1 Ci 192Ir pulsed-dose-rate source. The depth dependence of the film response was observed and a correction function was established. The suitability of the method was tested through measurement of the radial dose profile and radial dose function. The results were compared to Monte Carlo-simulated values according to the TG43 formalism. Monte Carlo simulations were performed separately for the beta and gamma source emissions, using the EGS4 code system, including the low-energy photon and electron transport optimization procedures. The beta source emission simulation showed that the beta dose contribution could be neglected and therefore the film-depth dependence could not be attributed to this part of the source radioactivity. The gamma source emission simulations included photon-spectra collection at several depths. The results showed a depth-dependent softening of the photon spectrum that can explain the film-energy dependence.

Monte Carlo calculation of radiation dose in CT examinations using phantom and patient tomographic models

By using a voxel-based Monte Carlo simulation technique, we developed and validated a method to calculate radiation-absorbed dose in the computed tomography (CT) examinations from the images of phantoms and patients. The ionising radiation transport was simulated using the EGS4 code system. The geometry of the X-ray beam (focus-to-axis distance, field of view, collimation, and primary and beam-shaper filtration) and the X-ray spectral distribution (HiSpeed LX/i) were included in the simulation. Each axial CT image was reduced to a 256 x 256 matrix and stacked in a volume. The patient images were segmented before the simulation of radiation transport by using four categories of materials, such as air, lung, muscle and bone. To test the voxel-based method, the values of the radiation dose derived from a simulated CT exposure were calculated and compared with those obtained from the measurements performed within the dosimetry phantoms. To complete the scope of the work, series of CT scans of the trunk of an anthropomorphic phantom and patients were simulated to calculate the average dose in each 1-cm-wide transverse slice (ADS). The comparison between the simulated and measured dose data for the CT indices showed a difference of <5% in all the cases. The estimated mean values of ADS from the chest, abdomen and pelvis of the anthropomorphic phantom were approximately 1.7-2 times the weighted CT dose index (CTDI(w)) value, whereas the mean ADS values for these anatomical areas were 1.3-2 times the CTDI(w) of patients. The voxel-based simulation method provided a technique for estimating the individual patient doses in the CT examinations.

X-ray fluorescence analysis in environmental radiological surveillance using HPGe detectors

X-ray fluorescence (XRF) has been proven to be a valuable tool for determining trace quantities of heavy metals, such as uranium and lead, in different types of samples. The present paper demonstrates the applicability of XRF spectrometry to measure the concentrations of these heavy metals in samples from natural ore and soil. The values of uranium concentrations in rock from the Peña Blanca uranium ore, in Chihuahua, México, were calculated for the purpose of precertifying the rock powders samples. The comparison with other techniques, such as inductively coupled plasma atomic emission spectrometry, atomic absorption spectrometry, alpha spectrometry and electron microscopy, was used to complete the precertification process, so that the sample powders may be used as secondary standards. The source-sample-detector geometry and the incident angle are the most important factors for obtaining low detection limits. The selected system uses a 57Co source of about 0.1 mCi to excite the K X-rays from uranium and lead. X-rays were recorded on a CANBERRA HPGe coaxial detector. The comparative results for two incident angles (90° and 180°) performed previously by other authors show that the best geometry is the backscattering geometry. In the present paper, using EGS4 code system with Monte Carlo simulation, it was possible to determine the location and distribution of background produced by the Compton edge in the optimized geometry. This procedure allowed to find the minimum detectable concentration of uranium and lead, which was experimentally calculated using standards. The possibility of performing in vivo measurements rapidly and easily, as well as the factors affecting accuracy and the minimum detectable concentration in several samples are also discussed.

Study of Compton scattering signals in single-sided imaging applications using Monte Carlo methods

Compton scatter imaging (CSI) is well establish one-sided imaging approach, where the image is formed using photons that are singly scattered in a region of the object. However, the measured data includes also photons that are multiply scattered and is also affected by the fact that incoming and outgoing beams are attenuated. In this paper, the EGS4 code system is used to gain some insight into the physics that constrains CSI. The yields of singly and multiply scattered photons, obtained with monochromatic and X-ray tube inspecting beams, were assessed for a variety of media and depths.

Precise Monte Carlo simulation of gamma-ray response functions for an NaI(Tl) detector

Three widely used methods to calculate the response functions for NaI(Tl) detectors were investigated. The methods were Berger–Seltzer's method (Nucl. Instrum. Methods 104 (1972) 317), general Monte Carlo (MC) programs, such as EGS4 (The EGS4 code system, Stanford Linear Accelerator Center, 1985) and MCNP4B (MCNP—a general Monte Carlo N-particle transport code, Los Alamos National Laboratory Report, LA-12625-M, 1997), and special MC programs. The pulse height spectra in a 3″×3″ NaI(Tl) detector due to several gamma-ray sources have been measured to verify the calculated results of these methods. The energies of the sources ranged from 0.4118 to 7.11 MeV. The spectra generated by Berger–Seltzer's method and the general MC programs did not agree well with the experimental data. PETRANS 1.0, the special MC program developed in house, was fairly accurate since it also considered the scintillation efficiency and the single escape peak shift.

Experimental study and Monte Carlo modeling of the Cherenkov effect

Studies realised at the Institute for Nuclear Research and Nuclear Energy (INRNE) particularly in cosmic ray detection and construction of Muonic Cherenkov Telescope at the South West University “Neofit Rilski” Blagoevgrad show the need to develop a theoretical model based on observed phenomena and to refinement of this for detection system optimisation. The Cherenkov effect was introduced in EGS4 code system. The first simulations realised in collaboration between the french and the bulgarian team were consecrated to different geometries of water tank in total reflection. An additional modeling of photons mean trajectory and the mean number of reflections in the tank were made. This simple model was compared with experimental data realised with 60Co gamma source, the telescope and the most efficient water tank. A trajectory simulation of Cherenkov photons in water tank was made. An efficiency estimation of the detector registration was calculated. The atmospheric model was introduced in EGS4 code and a comparison between CORSIKA5.62 and EGS4 codes was made.

Modelling and study of the Cherenkov effect

Studies at the Institute for Nuclear Research and Nuclear Energy particularly in regard to cosmic ray detection and construction of the Muonic Cherenkov telescope at the University of Blagoevgrad indicate a need for the development of a theoretical model based on observed phenomena and a refinement of this for detection system optimisation. This was introduced in the EGS4 code system. The first simulations consecrate on a number of different geometries of the water tank in total reflection. The model was compared with experimental data involving a 60Co gamma source and the telescope.

Enhancing the ratio of fluorescence to bremsstrahlung radiation in X-ray tube spectra

This paper describes techniques that can be used to improve the ratio of fluorescence to bremsstrahlung radiation ( F/ B) in X-ray tube spectra. Firstly, an extension of the EGS4 code system is used to evaluate the impact of the substrate in thin target applications, in terms of the yield of bremsstrahlung photons produced. The choice of materials to filter X-ray tube spectra, and its effect in the F/ B and the tube efficiency is discussed. The characteristics of spectra produced in transmission tubes with different target thicknesses, substrates and tube voltages are also presented.

Positronium rare annihilation experiment with a positron beam-line using permanent magnets

Positronium (Ps) is a bound state of an electron and a positron, which annihilates into photons. It is a suitable place to study the quantum electrodynamics (QED) since strong and weak interactions are negligibly small in the annihilation processes. Our experimental group has been measuring rare annihilation processes from Ps using a multi-γ-ray spectrometer called UNI which consists of 32 NaI(Tl) (3 in. φ ×4 in.) scintillation counters. We had detected the 4γ and 5γ annihilation processes for the first time in the world. In the experiment, ∼200 4-γ events and one 5-γ event have been collected, respectively, until today. Since these processes do not include any lower order of Feynman graphs, we can study α 7 and α 8 ( α : fine structure constant) processes almost purely, which is unexplored region of the QED higher order processes. We are planning to utilize a positron beam line using permanent magnets. By employing EGS4 code system for simulating electromagnetic processes and Runge–Kutta method for solving equations of motion, detection efficiency of our detector and transport efficiency of the beam line have been simulated.

Monte Carlo calculations of the ionization chamber wall correction factors for 192Ir and 60Co gamma rays and 250 kV x-rays for use in calibration of 192Ir HDR brachytherapy sources

As in the method for the calibration of 192Ir high-dose-rate (HDR) brachytherapy sources, the ionization chamber wall correction factor Aw is needed for 192Ir and 60Co gamma rays and 250 kV x-rays. This factor takes into account the variation in chamber response due to the attenuation of the photon beam in the chamber wall and build-up cap and the contribution of scattered photons. Monte Carlo calculations were performed using the EGS4 code system with the PRESTA algorithm, to calculate the Aw factor for 51 commercial ionization chambers and build-up caps exposed to the typical energy spectrum of 192Ir and 60Co gamma rays and 250 kV x-rays. The calculated Aw correction factors for 192Ir and 60Co sources and 250 kV x-rays agree very well to within 0.1% with published experimental data (the statistical uncertainty is less than 0.1% of the calculated correction factor value). For the 192Ir sources, Aw varies from 0.973 to 0.993 and for the 250 kV x-rays the minimum value of Aw for all chambers studied is 0.983. The calculated Aw correction factors can be used to calculate the air kerma calibration factor of HDR brachytherapy sources, when interpolative methods are considered, contributing to the reduction in the overall uncertainties in the calibration procedure.

Photon production using a low energy electron expansion of the EGS4 code system

An expansion of the EGS4 code system has been developed in which electron-atom inelastic collisions are modelled taking into account atomic bound effects. Additional models have been developed to deal with the ionisation of the K-shell and improved angular sampling of newly created bremsstrahlung photons. The incorporation of these features in EGS4 is described and the updated code is used to simulate photon generation in X-ray tubes. The accuracy of the new code and some of its possible applications are discussed.

Improvements of low-energy photon transport in EGS4

The treatment of low-energy (≤1 MeV) photon transport in the EGS4 code [Nelson, W.R., Hirayama, H. and Rogers, D.W.O., 1985. The EGS4 Code System SLAC-265 Stanford Linear Accelerator Center] has been improved. This also includes an improvement of the treatment of the electrons generated from low-energy photons and photons generated from low-energy electrons. The phenomena treated are: (i) linearly polarized photon scattering, (ii) Doppler broadening of Compton-scattered photons, (iii) L-X-ray and L-photoelectron production and (iv) electron impact ionization. We made measurements to verify the improved EGS4 code using monochromatized synchrotron radiation.

A Monte Carlo study of the quality dependence of diamond thermoluminescent dosimeters in radiotherapy beams

Monte Carlo simulations with the EGS4 code system have been performed to determine the quality dependence of diamond TLDs in photon beams ranging from 25 kV to 25 MV x-rays and also in megavoltage electron beams. It has been shown that diamond TLDs in the form of discs of thickness 0.3 mm and diameter 5.64 mm show no significant dependence on the incident energy in clinical electron beams when irradiated close to , but require an energy correction factor of compared with diamond TLDs irradiated in -rays. The correction factor increases with depth of irradiation and this effect is greater for thicker detectors. The Monte Carlo predicted sensitivity in x-ray beams is constant within 2.5% over the energy range 250 kV to 25 MV. However the sensitivity decreases by about 60% for 25 kV x-rays compared with -rays.

Electron transport: multiple and plural elastic scattering

We present a new approach to small angle multiple and plural scattering of electrons applicable for all step sizes in atomic and compound matter. Using the screened Rutherford law to describe single scattering and the momentum transfer as the “angular” variable, a simple analytical expression for the multiple scattering distribution is found. The parameters of this distribution are fixed by a χ 2-fit to the distribution calculated by the Monte Carlo technique. The validity of the small angle approximation is investigated. It is found, that at very large step sizes, the multiple scattering distribution becomes a Gaussian. The new approach is implemented in the EGS4 code system and tested in a variety of geometries. It is found, that the new multiple scattering subroutine is 2–3 times faster as the standard EGS4 one.

Effect of size and composition of the central electrode on the response of cylindrical ionization chambers in high-energy photon and electron beams

The authors present the Monte Carlo calculated central electrode correction factors for cylindrical ionization chambers used in high-energy photon and electron beams. The EGS4 code system with the application of a correlated sampling variance reduction technique has been employed in the calculation of the response of ionization chambers of different size and material. The results show that for an NE2571 chamber with a 1 mm diameter aluminium electrode the chamber response is increased by 0.6% in a 60Co beam and 0.4% in a 24 MV beam compared to a graphite electrode. In a 60Co beam the effect of an aluminium electrode compared with a graphic electrode increases with diameter from about 0.6% for 1 mm diameter to 3.4% for 5 mm diameter. The hollow aluminium electrode in an NE2561 chamber increases the chamber response by about 0.6% in a 60Co beam compared to an air electrode. For electron beams the electrode effect varies with incident electron energy and phantom depth.

A simple method to introduce K-edge sampling for compounds in the code EGS4 for X-ray element analysis

Abstract In the standard version of the EGS4 code system, the equivalent K-edge of a compound is assumed to be the weighted average of the K-edge of each element in the compound. This implies that the energy of both the fluorescent photon and the photoelectron is not correct, hence inaccurate EGS4 simulations of energy scoring in the compound material may be produced in the low-energy range (10–100 keV). In this report we show a technique that we used to overcome this problem. The same technique can also be used to introduce L-edge sampling in EGS4. An application of this sampling method to HgI 2 compound is shown.

Portable parallel programming in a Fortran environment

Experience using the Argonne-developed PARMACs macro package to implement a portable parallel programming environment is described. Fortran programs with intrinsic parallelism of coarse and medium granularity are easily converted to parallel programs which are portable among a number of commercially available parallel processors in the class of shared-memory bus-based and local-memory network based MIMD processors. The parallelism is implemented using standard UNIX tools and a small number of easily understood synchronization concepts (monitors and message-passing techniques) to construct and coordinate multiple cooperating processes on one or many processors. Bencmark results are presented for parallel computers such as the Alliant FX/8, the Encore MultiMax, the Sequent Balance, the Intel iPSC/2 Hypercube and a network of Sun 3 workstations. These parallel machines are typical MIMD types with 8–30 processors, such rated at 1–10 Mips processing power. The demonstration code used for this work is a Monte Carlo simulation of the response to photons of a “nearly realistic” lead, iron and plastic electromagnetic and hadronic calorimeter, using the EGS4 code system.

Shielding requirements for constant-potential diagnostic x-ray beams determined by a Monte Carlo calculation.

A Monte Carlo calculation has been performed to determine the transmission of broad constant-potential x-ray beams through Pb, concrete, gypsum wallboard, steel and plate glass. The EGS4 code system was used with a simple broad-beam geometric model to generate exposure transmission curves for published 70, 100, 120 and 140-kVcp x-ray spectra. These curves are compared to measured three-phase generated x-ray transmission data in the literature and found to be reasonable. For calculation ease the data are fit to an equation previously shown to describe such curves quite well. These calculated transmission data are then used to create three-phase shielding tables for Pb and concrete, as well as other materials not available in Report No. 49 of the NCRP.

Presta: The parameter reduced electron-step transport algorithm for electron monte carlo transport

A new electron transport algorithm for use with electron Monte Carlo transport codes is presented. Its components are: a path-length correction (PLC) algorithm which is based on the multiple scattering theory of Molière and which takes into account the differences between the straight path length and the total curved path length for each electron step; a lateral correlation algorithm (LCA) which takes into account lateral transport; and a boundary crossing algorithm (BCA) which ensures that electrons are transported accurately in the vicinity of interfaces. The algorithm has been implemented in the EGS4 code system and a variety of tests validating the algorithms are presented. In its standard configuration, use of this algorithm will allow the inexperienced user to obtain reliable results from the EGS Monte Carlo code without having to do a detailed study of the electron transport parameters. In many situations, substantial savings in computing time may be realized in comparison to the present EGS algorithm. The developments described in this report may be adapted to other electron transport codes where many of the same conclusions may be drawn.