
ORIGINAL ARTICLE 



Year : 2008  Volume
: 33
 Issue : 4  Page : 141146 

Inair fluence profiles and water depth dose for uncollimated electron beams
Abdelkader Toutaoui^{1}, Amar Nassim Aichouche^{2}, Kenza Adjidir^{2}, Ahmed Chafik Chami^{2}
^{1} Department de Physique Medicale, Centre de Recherche Nucleaire d Alger, 2Bd Frantz Fanon BP399 Alger RP, Algeria ^{2} Laboratoire de Sciences Nucleaires, Faculte de Physique, Universite des Sciences et de la Technologie Houari Boumedienne, BP 32 El Alia, Bab Ezzouar, Algiers, Algeria
Date of Submission  08Jul2008 
Date of Acceptance  26Oct2008 
Correspondence Address: Abdelkader Toutaoui Department de Physique Medicale, Centre de Recherche Nucleaire d Alger, 2Bd Frantz Fanon BP399 Alger RP Algiers Algeria
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/09716203.44473
Abstract   
Advanced electron beam dose calculation models for radiation treatment planning systems require the input of a phase space beam model to configure a clinical electron beam in a computer. This beam model is a distribution in position, energy, and direction of electrons and photons in a plane in front of the patient. The phase space beam model can be determined by Monte Carlo simulation of the treatment head or from a limited set of measurements. In the latter case, parameters of the electron phase space beam model are obtained by fitting measured to calculated dosimetric data. In the present work, data for air fluence profiles and water depth doses have been presented for electron beams without an applicator for a medical linear accelerator. These data are used to parameterize the electron phase space beam model to a Monte Carlo dose calculation module available in the first commercial (MDS Nordion, now Nucletron) Monte Carlo treatment planning for electron beams.
Keywords: Monte Carlo electron beam algorithm, phase space model, uncollimated electron beams
How to cite this article: Toutaoui A, Aichouche AN, Adjidir K, Chami AC. Inair fluence profiles and water depth dose for uncollimated electron beams. J Med Phys 2008;33:1416 
How to cite this URL: Toutaoui A, Aichouche AN, Adjidir K, Chami AC. Inair fluence profiles and water depth dose for uncollimated electron beams. J Med Phys [serial online] 2008 [cited 2019 Aug 25];33:1416. Available from: http://www.jmp.org.in/text.asp?2008/33/4/141/44473 
Introduction   
Advanced 3D electron beam dose calculation models such as the phasespace evolution model^{ [1],[2],[3]} and several 'macro' Monte Carlo^{ [4],[5],[6]} simulations require the input of an initial phase space (IPS) that describes a clinical electron beam. Such initial phase spaces describe the electron beam, the differential in space coordinates perpendicular to the beam axis, and the differential in the energy and angle.
An initial phase space is calculated using the simulation of electron transport through the head of a clinical linear accelerator and registers the electrons that enter the IPS plane. The EGS4/BEAM Monte Carlo code^{ [7]} is an implementation of this method. The results of this simulation can be parameterized by a multiplesource model^{ [8]} having the advantage of substantially less storage capacity.
As an alternative to the above method, simple models for the initial phase space have been proposed that are fully based on measurements.^{ [9]} Methods based on only a limited set of measured beam data have an advantage in clinical practice as they can be easily implemented in treatment planning systems.
In 2001, the first commercial Monte Carlo treatment planning for electron beams was released by MDS Nordion (now Nucletron), which has been implemented in the Dose Calculation Module (DCM) and can be accessed from the THERAPLAN PLUS™ or ONCENTRA treatment planning systems.
The implementation of the Monte Carlo calculation algorithm for electrons has two independent components: (i) a new, coupled MultiSource electron beam model^{ [10],[11],[12]} for electron transport through the linear accelerator treatment head, and (ii) a Monte Carlo electron and photon transport dose calculation code for inpatient dose calculation, the VMC++ Monte Carlo algorithm developed by Kawrakow and coworkers.^{ [5],[13],[14],[15]}
The coupled multisource beam model dealing with the electron transport through the accelerator head is based on an energy distribution of the electrons specified by a laterally invariant energy spectrum and smallangle fiveparameter parameterization. This describes the effective lateral and directional distributions of electrons.
The energy spectrum is determined by minimizing the difference between measured and calculated water depth doses from a large field depth dose measurement in an open electron beam (without an applicator). The parameters of the electron phase space beam model are obtained by fitting measured to calculated open field air fluence profiles for a set of rectangular field shapes defined by the block collimators.
In this work, measurement of air fluence profiles and water depth doses have been presented for electron beams without the applicator. These data were collected, for the radiation beam characterization of the THERAPLAN PLUS^{ TM} treatment planning system installed in the Radiotherapy Department of the Centre Pierre and Marie Curie, Algiers. The set of beam characterization dosimetric data includes, amongst other things, measured profiles and depth dose in air and water for each beam energy, both without and with the applicators on. The beams have been ''fitted'' by the vendor and the TPS is provided with the readytouse virtual treatment units.
Materials and Methods   
Incident beams and detectors
Measurements were performed in a Clinac 1800 linear accelerator (Varian) which produces nominal electron energies from 6 to 20 MeV. A special nonclinical mode (e.g., service mode) is used to perform the measurements for electron beams without an applicator, overriding interlocks of the machine. This was necessary for the measurements described in this paper, as well as for commissioning electron beams.^{ [1],[5]}
A commercially available radiation field analyzer (ScanditronixWellhofer RFA200, Uppsala, Sweden) was used to obtain all measurements in air and water. The relative dose was measured using a ptype silicon diode. A monitoring detector (similar diode) was used as a reference detector in air near the edge of the field to account for any effects of beam variations. Measurements were made in the continuous scanning mode with 1 mm resolution. The estimated overall relative uncertainties in our measurements were within 2%.
Air fluence measurements
Air fluence measurements were carried out in free geometry air without any electron applicator on the accelerator.
Inair measurements were performed using an electron diode without any buildup cap. The detector was located in a scanning device at the maximal height possible in an empty water phantom. Measurements were carried out along the central axis and in horizontal planes at two different sourcetodetector distances (SDD). For each electron beam energy, measurements were done for rectangular fields which consist of air fluence profiles along the inplane and crossplane directions for both sourcetodetector distances (SDD), as well as a depth dose measurement along the zaxis starting above the upper measurement plane (corresponding to the first SDD) and extending beyond the lower plane (corresponding to the second SDD).
For the rectangular fields, the setting of the uppermost jaw was fixed, defining an 8 cm slit field projected at isocenter plane. The lower jaw was varied to define widths of 8, 20, and 35 cm fields.
In addition to the above, measurements were also performed for a large square field of 35 × 35 cm^{ 2} area to measure the full penumbra.
For each field, measurements were carried out in two planes located 85 and 105 cm from the nominal target position, and these measurements covered the 9010% penumbra plus at least 4 cm. In addition to this, a scan along the central axis was acquired that was parallel with the radiation beam This scan starts 2 cm above the first SDD and extends 5 cm beyond the second SDD.
The spatial increment for the profile measurements in high gradient regions such as the field edges the scan step not exceed 0.1 cm and in low gradient regions, this increment was increased to 0.2 cm. The measurements along the central axis were performed with a spatial increment not exceeding 0.1 cm.
It was recommended that all inair profiles have a common normalization point that was not necessarily in absolute units. For the sake of convenience, all inair profiles were normalized to the signal from the 8 cm × 20 cm field on the central axes for the shorter SDD.
Dose measurements in water without applicator
The depth dose measurements were performed using a Scanditronix RFA200 dosimetry scanning system and a Scanditronix electron ptype diode detector. The diode signal was normalized, for each energy, at the normalization depth specified above. An open field (no applicator, 35 cm × 35 cm field size) depth dose measurement at SSD = 100 cm was acquired for each energy and without applicator. The measurements were normalized to 100.0 at the normalization depth and extend beyond the depth for the bremsstrahlung measurement listed in [Table 1]. These depth dose data will be used for deriving the energy spectrum for the phase space model.
Results   
The inair depth dose distributions along the z axis that have been presented in [Figure 1ac] for field sizes of 8 × 8 cm^{ 2} , 20 × 8 cm^{ 2} , 35 × 8 cm^{ 2} , and 35 × 35 cm^{ 2} for electron beam energies 6, 12, and 20 MeV respectively. For all electron beam energies, it is seen that the central axis depth dose for a field size of 8 × 8 cm^{ 2 } presents values greater than those of the depth dose curves of other field sizes.
The difference between the values of the depth dose curve for the 8 × 8 cm^{ 2} field and those of the depth dose curves for the 20 × 8 cm^{ 2} , 35 × 8 cm^{ 2} , and 35 × 35 cm^{ 2} (presented in [Figure 1ac]) fields at the point nearest to the treatment head (located at 83 cm SSD), is 20% for the electron beams of energy 6 and 9 MeV, of 15% for the beam of 12 MeV, of 10% for the beam of 16 MeV, and 6% for the beam of 20 MeV. These differences are reduced to approximately 5% at the end of the curve for all electron beam energies.
The values of the inair depth dose curves for the 20 × 8 cm^{ 2} , 35 × 8 cm^{ 2} , and 35 × 35 cm^{ 2} fields [Figure 1ac] are close between them. When the electron beam energy increases, these inair depth dose curves approach each other.
From the results presented in [Figure 1], it appears clearly that the slope of the depth dose decreases when the energy of the electron beam increases. With the increase of the beam energy, the scattering of the electrons in the air decreases.
[Figure 2] shows inair fluence profiles in the inplane direction measured for field sizes of 8 × 8 cm^{ 2} , 20 × 8 cm^{ 2} , and 35 × 8 cm^{ 2} rectangular fields, at two distances, 85 and 105 cm, from the source of the beam, for electron beam energies of 6, 12, and 20 MeV. In [Figure 2], it is also seen that the values of the inair fluence on the central axis for the field size of 8 × 8 cm^{ 2 } are larger than the inair fluence values for field sizes of 20 × 8 cm^{ 2} and 35 × 8 cm^{ 2} . [Table 2] shows the ratios of the dosimeter readings at the central axis at the distance of measurement of 85 cm, for the different field sizes, normalized to that of 8 × 8 cm^{ 2} for various electron beam energies. [Figure 2] illustrates that the inair fluence profiles do not exhibit a flat area in the central part of the field for the field size of 8 × 8 cm^{ 2 } and show a narrow flat area for the other field sizes in contrast to what could be observed for the inair fluence profiles measured for collimated beams.
[Figure 3] illustrates the profiles of inair fluence measured in the crossplane direction for the same field sizes, same electron beam energies, and at the same depths. It can be seen clearly that when the energy of the beam increases, the flat area of inair fluence profiles widens at the level of the central axis. It is seen that the widening of the profiles of inair fluence decreases when the energy of the beam increases.
[Figure 4] shows the measured depth doses in water for an open field of 35 cm × 35 cm area at 100 cm SSD, for 6, 9, 12, 16, and 20 MeV without an applicator.
The depth dose in water of uncollimated electron beams show a depth of the maximal dose and a practical range larger than those of the collimated beams.
Discussion   
For the inair depth dose curves, as shown in [Figure 1ac], the central axis depth dose for a field size of 8 × 8 cm^{ 2 } presents values greater than those of the depth dose curves of the other field sizes. This is mainly due to the contribution of the electrons scattered by the jaws of the collimators of photons. When the opening of these jaws increases beyond a certain size, the larger distance the scattered electrons have to traverse does not enable them to reach the central axis. The maximal difference is observed at the level of the point nearest to the treatment head. This variation tends to decrease with the distance of the point of measurement to the electron source for a given electron beam energy.
It was noticed from [Figure 2ac] that the inair fluence profiles do not exhibit or have a narrow flat area in the central part of the field. This is due to the lack of electrons scattered by the walls of the applicator. This justifies the parameterization of the phase space beam model from these profiles of fluence which represent the contribution of direct electrons and to a lesser degree, the electrons scattered by the jaws of the photon collimators.
The widening of the flat area of the inair fluence profiles at the level of the central axis, with the increase in electron beam energy can be observed in [Figure 3ac]. This widening can be explained by the fact that for high energy electron beams, the contribution of the electrons scattered by the jaws of photon collimators becomes more important. Their range becomes sufficient and this enables them to reach the central part of the beam.
The depth dose curves in water measured for the 35 × 35 cm^{ 2} field size for an electron beam without applicator will be used in the radiation data characterization process for deriving the energy spectra for unscatterd electrons as a part of the electron phase space beam model.
Summary and Conclusion   
This article presents a set of measured inair fluence and depth dose in water for an uncollimated electron beam, as a part of the dosimetric data set for the commissioning of a Monte Carlobased treatment planning for electron beams. These data are dedicated to the parametrization of the phase space beam model.
The choice of these data is governed by the requirement of the dose calculation algorithm to model the direct electron and the electrons scattered by the walls of the applicator as well as for the determination of the energy spectrum of the direct beam.
The phase space beam model parameters are derived as part of the treatment unit characterization process by optimizing calculated data towards measured open field data (no electron applicator mounted on the treatment head).
The energy spectrum is determined by minimizing the difference between measured and calculated water depth doses.
These data were sent to the manufacturer for parameterization beams and then built in into the TPS which is in the course of commissioning.
Acknowledgment   
We make a point of thanking Mr. R. Benali who allowed us to take measurements without an applicator.
References   
1.  Korevaar EW, Dabrowskiz R, Janssen JJ, Storchi PR, Huizenga H. Phase space evolution distribution functions for high energy electron beams. Phys Med Biol 1996;39:135166. 
2.  Janssen JJ, Riedeman DE, MorawskaKaczynska M, Storchi PR, Huizenga H. Numerical calculation of energy deposition by highenergy electron beams: III, Threedimensional heterogeneous media. Phys Med Biol 1996;39:135166. 
3.  Janssen JJ, Korevaar EW, Storchi PR, Huizenga H. Numerical calculation of energy deposition by highenergy electron beams: IIIB, Improvements to the 6D phase space evolution model. Phys Med Biol 1997;42:14419. 
4.  Neuenschwander H, Born E J. A macro Monte Carlo method for electron beam dose calculations. Phys Med Biol 1992;37:10725. 
5.  Kawrakow I, Fippel M, Friedrich K. 3D electron dose calculation using a voxelbased Monte Carlo algorithm (VMC). Med Phys 1996;23:44557. 
6.  Sempau J, Wilderman SJ, Bielajew AF. DPM: A fast, accurate Monte Carlo code optimized for photon and electron radiotherapy treatment planning dose calculations. Phys Med Biol 2000;45:226391. 
7.  Rogers DW, Faddegon BA, Ding GX, Ma CM, We J. BEAM: A Monte Carlo code to simulate radiotherapy treatment units. Med Phys 1995;22:50324. 
8.  Ma CM, Faddegon BA, Rogers DW, Mackie TR. Accurate characterization of the Monte Carlo calculated electron beams for radiotherapy. Med Phys 1997;24:40117. 
9.  Janssen JJ, Korevaar EW, Van Battum LJ, Storchi PR, Huizenga H. A model to determine the initial phase space of a clinical electron beam from measured beam data. Phys Med Biol 2001;46:26986. 
10.  Ahnesjφ A, Traneus E, Åsell M, SheikhBagheri D. Electron Beam Characterization for Monte Carlo Treatment Planning. Proceedings WC2000. Chicago, USA: 2000. 
11.  Ahnesjφ A, Traneus E, Åsell M. Generation of phase space for electron beam Monte Carlo treatment planning. Radiother Oncol 2000;56:S79. 
12.  Ahnesjφ A, Traneus E, Åsell M. Application and verification of a coupled multisource electron beam model for Monte Carlo based treatment planning. Radiother Oncol 2001;61:S102. 
13.  Kawrakow I, Fippel M. VMC++, a fast MC algorithm for radiation treatment planning XIIIth Int'l Conf. on The Use of Computers in Radiotherapy. Heidelberg: SpringerVerlag, Heidelberg, Germany: 2000. 
14.  Kawrakow I, Fippel M. Investigation of variance reduction techniques for Monte Carlo photon dose calculation using XVMC. Phys Med Biol 2000;45:216383. 
15.  Kawrakow I. VMC++, Electron and Photon Monte Carlo Calculations Optimized for Radiation Treatment Planning Advanced Monte Carlo for Radiation Physics, Particle Transport Simulation and Applications: Proceedings of the Monte Carlo 2000 Conference. Lisbon:Springer, Berlin, Germany; 2001. 
[Figure 1ac], [Figure 1], [Figure 2], [Figure 2ac], [Figure 3], [Figure 3ac], [Figure 4]
[Table 1], [Table 2]
