Radiation Protection Dosimetry Advance Access published December 2, 2013 Radiation Protection Dosimetry (2013), pp. 1–7
doi:10.1093/rpd/nct315
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS, TLDS AND IONISATION CHAMBER AND COMPARISON WITH MONTE CARLO SIMULATION
*Corresponding author: [email protected] Received 12 September 2013; revised 31 October 2013; accepted 6 November 2013 Estimation of the surface dose is very important for patients undergoing radiation therapy. The purpose of this study is to investigate the dose at the surface of a water phantom at a depth of 0.007 cm as recommended by the International Commission on Radiological Protection and International Commission on Radiation Units and Measurement with radiochromic films (RFs), thermoluminescent dosemeters and an ionisation chamber in a 6-MV photon beam. The results were compared with the theoretical calculation using Monte Carlo (MC) simulation software (MCNP5, BEAMnrc and DOSXYZnrc). The RF was calibrated by placing the films at a depth of maximum dose (dmax) in a solid water phantom and exposing it to doses from 0 to 500 cGy. The films were scanned using a transmission high-resolution HP scanner. The optical density of the film was obtained from the red component of the RGB images using ImageJ software. The per cent surface dose (PSD) and percentage depth dose (PDD) curve were obtained by placing film pieces at the surface and at different depths in the solid water phantom. TLDs were placed at a depth of 10 cm in a solid water phantom for calibration. Then the TLDs were placed at different depths in the water phantom and were exposed to obtain the PDD. The obtained PSD and PDD values were compared with those obtained using a cylindrical ionisation chamber. The PSD was also determined using Monte Carlo simulation of a LINAC 6-MV photon beam. The extrapolation method was used to determine the PSD for all measurements. The PSD was 15.0+ + 3.6 % for RF. The TLD measurement of + 3.0 %. The theorthe PSD was 16.0+ + 5.0 %. The (0.6 cm3) cylindrical ionisation chamber measurement of the PSD was 50.0+ etical calculation using MCNP5 and DOSXYZnrc yielded a PSD of 15.0+ + 2.0 % and 15.7+ + 2.2 %. In this study, good agreement between PSD measurements was observed using RF and TLDs with the Monte Carlo calculation. However, the cylindrical chamber measurement yielded an overestimate of the PSD. This is probably due to the ionisation chamber calibration factor that is only valid in charged particle equilibrium condition, which is not achieved at the surface in the build-up region.
INTRODUCTION The absorbed dose at the surface of a patient or phantom irradiated with a beam of megavoltage X rays arises from electrons generated in air above the phantom, from electron backscatter within the phantom, and from electrons generated by any solid material in the beam. So, the skin-sparing effect for high-energy gamma- and X-ray photons may be reduced or even lost(1). The relevant dose specification depth depends on the biological effect considered. According to the International Commission on Radiological Protection(2) and International Commission on Radiation Units and Measurement(3) the skin dose should be measured at 0.007 cm, and this depth generally corresponds to the interface between the epidermis and dermis layers of the skin. The surface dose can be defined as the energy deposited within an infinitesimally small mass of tissue at the surface of the phantom(1). Avariety of dosemeters can be used, such as fixed-separation parallel-plate chambers, radiochromic films (RFs) and TLDs. However, there is no dosemeter that has an infinitesimally small sensitive volume, and the surface dose by
definition is inherently difficult to measure. Most treatment planning systems (TPS) fail to accurately predict the dose at the skin of the patient because the skin cannot be defined in the computer tomography slices accurately and also because the voxel size used by the TPS is much larger than the skin depth (0.007 cm)(4). But the effective point of measurements for RFs and TLDs are close to the skin depth. So, they can be used for investigating the per cent surface dose (PSD) defined as the central axis skin dose relative to maximum dose (Dmax) at a depth of 0.007 cm for a 10` 10 cm2 field size. In the build-up region, the charged particle equilibrium does not exist and therefore an accurate measurement of PSD using an ionisation chamber cannot be obtained. Therefore, the choice of the measurement device is of great importance in PSD measurements. The introduction of RFs has reduced some of the problems encountered with other conventional radiation dosemeters. The high spatial resolution of RFs makes them ideal for the measurement of dose distributions in regions of high-dose gradients in radiation fields. There are three recent new RF models the
# The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
Saleh Alashrah1,2, Sivamany Kandaiya2, Nabil Maalej3 and A. El-Taher1,4,* 1 Department of Physics, Qassim University, Qassim, Saudi Arabia 2 Universiti Sains Malaysia, Penang, Malaysia 3 Department of Physics, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia 4 Physics Department, Faculty of Science, Al-Azher University, Assuit, Egypt
S. ALASHRAH ET AL.
I0 OD ¼ Log I where I0 is the light intensity with no film present and I the light intensity after passing through the film. The OD is usually a non-linear response to dose and hence requires calibration. The film is normally calibrated using a calibrated ionisation chamber. This approach has been able to provide useful dosimetric information. The uncertainties in dose measurements based on any film dosimetry system results from the uncertainty due to the effective point of measurements, non-uniform thickness of the film’s sensitive layer, the densitometer used to measure the OD and non-uniform radiation field during exposure(7). Thermoluminescent dosemeters (TLD) are used clinically due to the following advantages: (i) wide useful dose range, (ii) small physical size, (iii) no need for high voltage or cables and (iv) tissue equivalence (LiF) for most radiation types(8). TLDs can be particularly used for the absorbed dose measurements in areas where dose measurement is difficult and not as part of a routine verification procedure. The error in the TLD reading was found to be within 5 %(8). Unlike RF the LiF TLD can be annealed and re-used many times. In the surface region, charged particle equilibrium does not exist, as in all transition zones between two different media. This will cause perturbation effects in ionisation chambers. A PTW 30013 0.6-cm3 ionisation chamber was used because of its high accuracy in the charged particle equilibrium region. The chambers of fixed volume will tend to indicate a larger dose
in the build-up region than the actual dose(9). This increase in the measured dose is mainly a result of secondary electrons scattered into the chamber volume from the walls of the chamber. For this reason, it is recommended to use an extrapolation chamber to estimate the PSD(10, 11). By measuring the ionisation per unit volume as a function of electrode spacing, the superficial dose can be estimated by extrapolating the ionisation curves to zero electrode spacing. The perturbation effects in the ionisation chamber used for build-up measurements have been studied(12). The fluence perturbation due to electrons emitted through the side walls that cause an overestimation of the surface dose, have been thoroughly investigated by measurements using film, extrapolation chambers and by calculations(12). The perturbation of the electron fluence in the build-up region is due to lack of equilibrium in the transport of different categories of electrons contributing to the ionisation in the chamber. The main contribution to the ionisation, especially for small plate separations, is due to electrons coming from the air and treatment head. These electrons may then be backscattered. Photons hitting the chamber may emit electrons in the front electrode, collector and through the side walls. These electrons may then be scattered into the chamber volume. Electrons hitting the phantom close to the chamber may be scattered into the chamber. A small contribution is also obtained from electrons produced by photon interactions in the chamber(12). In principle, Monte Carlo simulations have a high spatial resolution when the used voxel size is small. The Monte Carlo technique involves using known probability distributions that govern the physical interactions of photons and electrons in various materials to simulate random trajectories of individual particles. By keeping track of processes of interest for a large number of histories, information regarding the average quantities and their correlated distributions can be obtained as well as the statistical fluctuations of specific events. The use of MC methods in radiation therapy physics has increased over the past few decades as high-energy photon and electron beams are being used in radiotherapy. It is necessary to account for the electron transport for dosimetry and treatment planning systems(13). In 1995 the EGS4 code was released(14) and is now upgraded to EGSnrc (electron gamma shower)(15), which is used for simulating electron and photon beams. EGSnrc has been employed in the medical physics to study problems in radiation dosimetry and radiation therapy. In addition, it is used in diagnostic imaging, design and characterisation of medical devices and radiation protection. This code allows easy modelling of a series of linear accelerators (Siemens, Elekta and Varian linacs and NRC research linac) and has led to publications of many of papers dealing with photon and electron beams from accelerators(16, 17).
Page 2 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
XR-T, HS and EBT. They have been introduced by International Specialty Products (Wayne, NJ, USA). In this experiment, the RF model HS was used. It has a single sensitive layer designed for two-dimensional dose measurements in high-energy photon beams (.1 MeV)(1). RF uses radiochemical processes to impart a change in the optical absorbance of the film for specific wavelengths of light. It can be used to measure very high doses associated with industrial irradiators and accelerators. The films are affected by exposure to bright light, temperature and humidity. Furthermore, ultraviolet light produces an unwanted colouration of the film. Therefore, appropriate handling and storage conditions are required. The films should thus be stored in an opaque container and only taken out for experiments and readout (5). The optical density (OD) of the films after exposure is preserved for some time. The disadvantage of these films is that they can only be used once and should be scanned at a defined time after exposure preferably 2 d after the exposure(6). The reduction in light passing through the film is a measure of its ‘blackness’ or ‘optical density’ expressed by
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS
MATERIALS AND METHODS Solid water phantom (RW3)
RF measurements The HS GafChromicw film model (International Specialty Products, Wayne, NJD) has been developed as a more sensitive and uniform alternative to the GafChromicw MD-55 film. It was specifically designed for the measurement of absorbed dose in high-energy photon and electron beams (.1 MeV). According to the manufacturer, the HS model covers a dose range from 1 to 50 Gy. The HS film consists of a single active layer sandwiched between two sheets of clear, transparent polyester, each with a thickness of 97 mm and a density of 1.38 g cm23 (Figure 1). The HS active layer has a thickness of 40 mm. The films were placed at dmax in a solid water phantom (30` 30` 20 cm3) and in the centre of a 10` 10 cm2 field size of 6-MV photon beam (from a
Dfit ¼ a þ b netOD þ c netODn
ð1Þ
The third term in Equation (1) is introduced to account for the non-linear dose response in the highdose region close to the saturation level for a given film dosimetry system. In order to predict the error in measurements of an unknown dose while using the calibration curve for each dosimetry system, the expression for error propagation was used:
s2y ¼
X @y2 i
@x
:s2xi
ð2Þ
where sy is the total estimated uncertainty (standard deviation) for a dose determined and sxi (i ¼ 1, 2, 3) is the standard deviations for the net OD and the two fitting parameters b and c. From the last equations, it follows:
Figure 1. Structure and dimension of HS RF model used in this study.
sPDD
Page 3 of 7
ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi s 2 s 2 max ¼ þ D Dmax
ð3Þ
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
However, water phantom measurements are difficult to set up and many detectors in radiation therapy are not waterproof(18). Solid water equivalent phantoms such as RMI457 solid water, plastic water, RW3 solid phantom (PTW Freiburg) and Perspex(19) are now available. The solid water phantom has similar properties such as physical density, relative electron density and effective atomic number as well as absorption and scattering of radiation to those of water(18). The electron density of the medium is important as the Compton scatter is a dominant process at highenergy photon beams (MV). Hill et al.(18) studied RMI457, plastic water, RW3 solid phantom and perspex using the source 99Tcm. The maximum difference in the absorbed dose results was 4 % between the phantoms and water. Slabs of the solid water phantom (RW3) were used in this study. The solid water phantom is an epoxy resin-based solid substitute of water. Its absorption and scattering properties is within 1.0 % of water. The physical density ranges from 1.039 to 1.049 g cm23. These characteristics allow the solid water phantom to be used in high energy (radiation therapy)) measurements(20). The dimension of the phantom is 30` 30 cm2 and the slabs thickness varied from 0.1 to 1 cm.
Varian 21 EX LINAC) as shown in Figure 1. The source-to-surface distance (SSD) was 100 cm. The films (3.5` 3 cm2) were exposed at different doses from 0 to 500 cGy for calibration. For each exposure, the film pieces were removed from their light-protecting envelope and irradiated. After 2 d from the exposure, an HP Scanjet 4890 scanner was used to scan the films. The scanner has a film holder for positioning the film pieces. Once the scanner is turned on, it is important to allow the scanner to warm-up and allows its temperature to stabilise. The exposed films were scanned six times and averaged to reduce the error in measurements. After the pieces of the films were scanned, the image J (http://rsb.info.nih.gov) program was used to extract the red component of the red– green –blue (RGB) image as the absorption spectrum of the RF exhibits a maximum response in the red region of the visible spectrum. Subsequently, the net OD of the irradiated film pieces was determined by subtracting the OD of an unexposed film. The relation between the doses and optical densities were plotted to obtain a calibration curve. Matlab 7.0.1 program was used to obtain the best fit curve of the calibration. In order to find the most suitable fit of the dose as a function of OD, the following criteria were used: (1) the fit function has to be monotonically increasing; (2) the fit function has to go through zero and finally; (3) chose the function that gives the minimum relative uncertainty for the fitting parameters. Based on these criteria, a family of fitting functions of the form was chosen:
S. ALASHRAH ET AL.
Ionisation chamber measurements The ionisation chamber used in the present work is a PTW 0.6-cm3 Farmer chamber (type 30013). The cavity length is 23 mm with the wall materials made of graphite and polymethyl methacrylate. Measurements with the ionisation chamber were performed in a water phantom (30` 30` 30 cm3) at different depths (0.5, 1, 1.5, . . . , 20 cm). All experiments were performed with a dose of 1 Gy at dmax, SSD ¼ 100 cm and 10` 10 cm2 field size at the surface of water phantom. The chamber was positioned with its axis perpendicular to the beam axis, as recommended by the manufacturer for reference dosimetry. The ionisation readings were taken at the effective point of ionisation chamber, i.e. the chamber was shifted towards the surface by 0.6` radius of the cavity to obtain the PDD curve as recommended by TG51(21) and TRS 398(22). Monte Carlo simulation
TLD Measurements ` 3.2 mm ` 0.89 Thirty-six TLD-100 chips (3.2 mm mm) were placed in a solid water phantom (30` 30` 20 cm3) at a depth of 10 cm using a TLD holder made of Perspex. The holder had holes of the same size as the TLDs to reduce the effect of air gaps. The TLDs were irradiated with a 6-MV photon beam at 10` 10 cm2 field size on the phantom surface with a 100-cm SSD. After the TLDs were exposed, they were read using a TLD reader (system 4000 HARSHAW). Each TLD has a characteristic calibration factor value which relates its response to dose. Subsequently, the surface dose was measured using the TLDs. The TLDs were handled with vacuum tweezers using a plastic nozzle to avoid scratching the surface of the chips. All experiments were performed at a dose of 1 Gy at d max.to the supra-linearity region of LiF(1). The effective point of measurement was chosen at the centre of the TLD chip. TLDs were placed at different depths (0, 1, 2, 3, 4 . . . , 190 mm) in the phantom and were exposed using 100 MU, 6-MV photon beam, 10` 10 cm2 field size at the phantom surface, and an SSD of 100 cm as shown in Figure 3.
MCNP5 and BEAMnrc were used to calculate the energy absorbed in the build-up and decay regions of the 6-MV photon beam irradiating perpendicularly to the water phantom. A photon beam model was simulated using the geometry and materials composition of a Varian 2100EX (Varian Medical Systems, Palo Alto, CA, USA). A schematic representation of the LINAC head and its components is shown in Figure 4. The LINAC head components, including the target, primary collimator, flattening filter and secondary collimator jaws, were simulated based on manufacturerprovided information. The SSD was simulated to be at 100 cm and the field size on the surface of water phantom to be 10` 10 cm2. In most calculations, a photon transport cut-off of 0.01 MeV and an electron kinetic energy transport cut-off of 0.10 MeV for MCNP5 were used but the cut-off energy for electron and photon was 0.7 and 0.01 MeV, respectively, for EGSnrc/BEAMnrc and EGSnrc/DOSXYZnrc. The energy of the electron source was 6 MeV and the Gaussian electron distribution was used in the simulation(23). For BEAMnrc, the ISOURC¼19 was used as a source with 6 MeV and the full width at half
Figure 2. The RF measurement set-up.
Figure 3. The TLD measurements set-up.
Page 4 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
where s and smax are the uncertainties in dose D at an depth d and Dmax at an depth dmax. The films were placed at different depths (0, 1, 2, 3, 4, 5, 7..., 190 mm) as shown in Figure 2. They were adjusted to be in the centre and perpendicular to the beam with a 10` 10 cm2 field size at the phantom surface. The SSD was 100 cm and the gantry angle was 08. The dose of 3 Gy was delivered at the depth of a maximum dose. For each exposure, the exposed film was removed and another film was placed in a new position. The HP4890 scanner was used for scanning the pieces of film and ImageJ program was used for estimating the average OD. The net OD were converted to dose by using the calibration curve and the dose was normalised to the maximum dose at dmax to obtain the PDD (PDD ¼ D/Dmax). Finally, the error in the dose was calculated.
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS
Figure 4. The geometry of the head of LINAC (2100 EX) (a) using BEAMnrc and (b) for MCNP5 simulation.
maximum (FWHM) of the distribution of the source in x and y directions was 0.15 cm. The number of particles history was 80` 106 for MCNP5 and it was 1` 109 for BEAMnrc and DOSXYZnrc. The detectors were simulated to be at the centre of the phantom and perpendicular to the beam. These detectors were cubic with dimensions 2` 2` 0.02 cm3 for MCNP5. They were distributed at different depths (1, 1.3, 1.5, 1.7, 1.9, 2, . . . 190 mm). The DOSXYZnrc was used to simulate the phantom (50` 50` 50 cm3). The detector size was 0.2` 0.2` 0.2 cm3 in the build-up region. All the software (MCNP5, BEAMnrc and DOSXYZnrc) ran on a window vista PC (Intel core TM 2 Quad Process, 2.6 GHz, 500 GB hard drive, 4 GB memory).
RESULTS AND DISCUSSION The calibration curve for the HS RF model is shown in Figure 5. The best fit curve of dose versus the measured net OD is determined using Equation (1). The best fit was obtained with n ¼ 5. Using MTLAB 7.0.1 the best values of the parameters b and c with the corresponding uncertainties were determined.
Figure 6. Per cent depth dose curves for a 6-MV photon beam at the build-up region using RF, TLD-100, ionisation chamber, Monte Carlo simulation using MCNP5 and EGSnrc/DOSXYZnrc.
To estimate the clinically relevant skin dose, the effective points of measurement for the dosemeters must be known. For the ionisation chamber, the effective point is at a distance equal to 0.6 r cavity from the geometric centre where r cavity is the radius of the cavity of the cylindrical ionisation. On the other hand, for RFs and TLDs, the effective point of measurement was defined at the midpoint of the sensitive layer. For Monte Carlo results, the effective point measurement was defined at the geometric centre of the detector. Figure 6 shows the PDD curve in the build-up region using the RF, TLD, ionisation chamber, DOSXYZnrc and MCNP5 using SSD ¼ 100 cm. The difference in the results is in the build-up region between TLDs and EGSnrc or MCNP5 was within +5 %. However, the difference between RF and Monte Carlo simulation was within 10 %. Other groups have reported similar differences between the
Page 5 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
Figure 5. Calibration curve for HS RF with best fitting (D ¼ b` net OD þ c` net OD5, where b ¼ 4.956+0.43 and c ¼ 2.198` 1028 + 1.38` 1028 and R 2 ¼ 0.9982).
S. ALASHRAH ET AL. Table 1. The difference in the results for RF, TLD-100 and MCNP5 used in this study for dose at a depth of 20 cm. Dosemeter or calculation method PTW 30013 Farmer ionization chamber HS radiochromic film TLD-100 (3.2` 3.2` 0.89) MCNP5 BEAMnrc
PDD (d ¼ 20 cm) (%) 38 38 40 40 40
CONCLUSION measured and calculated doses in the build-up region reaching up to 10 %(24). The Monte Carlo [MCNP5 or EGSnrc (BEAMnrc/DOSXYZnrc)] simulation with a small pixel size has a high resolution, and no electronic equilibrium is required and provides accurate results. Consequently, the Monte Carlo simulation has been used as a gold standard for dose calculation. The PTW 30013 Farmer ionization chamber readings agree with MCNP5 and DOSXYZnrc at a depth of .4 mm. For depths shallower than 4 mm, there is a large difference between the chamber reading and MC simulation results. The skin depth was assumed to be at the reference point of 70 mm; therefore, an extrapolation method needs to be used to estimate the percentage skin dose. According to HS RF, the percentage skin dose is 15.0+3.6 % and it is 16.0+5.0 % for TLD-100. These results agree with MCNP5 (15.0+2.0 %) and DOSXYZnrc (15.7+2.2 %) results. When the TLDs were used to measure the PSD, their results agreed with the published results(25). Devic et al.(1) reported a PSD measurement of 16 % using the Attix parallel plate ionisation chamber. On the other hand, the PSD measurement obtained in this study using cylindrical (0.6 cm3) ionisation chamber is 50.0+3.0 %. Therefore, for percentage skin measurements, a thin dosemeter has to be used. A cylindrical chamber reading gives an overestimate of the PSD due to the presence of partial charged particle equilibrium. Figure 7 shows the measured PDD curve in the region after dmax using RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5. There is a very good agreement between RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5 results. For instance, the PDD results at a depth of 20 cm are shown in Table 1. Based on the results of Figure 7 and Table 1, the maximum difference in the PDD at a depth of 20 cm is ,4 %. The data show that both RF and TLD gave a better estimate of PSD than that from the cylindrical ionisation chamber. A parallel-plate extrapolation chamber than a cylindrical chamber will give a more accurate estimate of PSD.
The skin dose was investigated at a depth of 0.007 cm experimentally and by Monte Carlo simulation. Experimental measurements were done using RFs, TLD-100 and a cylindrical ionisation chamber (0.6 cm3) in a solid water phantom. The PSD was 15.0+3.6 and 16.0+5.0 % for RF and TLD-100, respectively. Ionisation chamber measurements were extrapolated to the surface and yielded a PSD of 50.0+3.0 %. The Monte Carlo simulation does not require electronic equilibrium and has the advantage of high resolution due to small voxel size and provides accurate dose calculation. Consequently, the Monte Carlo simulation has been used as a gold standard for dosimetry. The PSDs using MCNP5, EGSnrc/ BEAMnrc and EGSnrc/DOSXYZnrc were 15.0+2.0 and 15.7+2.2 %. Therefore, there is a good agreement between RF and TLD measurement with Monte Carlo results. However, the PSD measurement using an ionisation chamber was 50.0+3.0 %, very different from other Monte Carlo results. The reason for the overestimate of dose measurement using the ionisation chamber in the build-up region is mainly due the contribution of secondary electrons scattered into the chamber volume. The measurements obtained in this study have also shown that there is an agreement between RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5 in the region after dmax. The maximum percentage difference between the dosemeters was ,4 % at a depth of 20 cm. Extrapolating the cylindrical chamber measurements to the surface yields an overestimate of the PSD. To obtain an accurate measurement of the skin dose which agrees with the Monte Carlo calculation, thin dosemeters should be used such as RF and TLD. The main advantage of RF is small size and high resolution. However, RF requires calibration and the OD response is not linear with dose. The TLD response is linear with dose but requires meticulous handling during calibration, irradiation and reading and are usually larger in thickness compared with RF. Both RF and TLD do not provide immediate reading and require post-irradiation processing.
Page 6 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
Figure 7. The PDD curves in the decay region using RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5.
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS
ACKNOWLEDGEMENTS The authors would like to thank Mr Hassan Al-Gamdi from Dhahran Medical Centre (Saudi ARAMCO) and to Dr Tarak from King Faisal Specialist Hospital and Research Centre for their help. REFERENCES
Page 7 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
1. Devic, S., Seuntjens, J., Abdel-Rahman, W., Evans, M., Olivares, M., Podgorsak, E., Vuong, T. and Soares, C. G. Accurate skin dose measurements using radiochromic film in clinical applications. Med. Phys. 33, 1116 (2006). 2. ICRP. Recommendations of the International Commission on Radiological Protection. ICRP Publication 60. ICRP 21(1–3). Pergamon (1991). 3. ICRU. Determination of dose equivalents resulting from external radiation sources. Report No. 39. International Commission on Radiation Units and Measurement (1985). 4. Stathakis, S., Paskalev, J., Li, K., Yang, J., Wang, L. and Ma, C. Ultra-thin TLDs for skin dose determination in high energy photon beams. Phys. Med. Biol. 51, 3549 (2006). 5. Butson, M. J., Yu, P. K. N., Cheung, T. and Metcalfe, P. Radiochromic film for medical radiation dosimetry. Mater. Sci. Eng. R Rep. 41, 61– 120 (2003). 6. Niroomand-Rad, A., Blackwell, C. R., Coursey, B. M., Gall, K. P., Galvin, J. M., McLaughlin, W. L., Meigooni, A. S., Nath, R., Rodgers, J. E. and Soares, C. G. Radiochromic film dosimetry: recommendations of AAPM radiation therapy committee task group 55. Med. Phys. 25, 2093 (1998). 7. Devic, S., Seuntjens, J., Sham, E., Podgorsak, E. B., Schmidtlein, C. R., Kirov, A. S. and Soares, C. G. Precise radiochromic film dosimetry using a flat-bed document scanner. Med. Phys. 32, 2245– 2253 (2005). 8. Aschan, C. Applicability of thermoluminescent dosimeters in x-ray organ dose determination and in the dosimetry of system and boron neutron capture radiotherapy. Helsingin Yliopiston Verkkojulkaisut (1999). 9. Pena, J., Sa´nchez-Doblado, F., Capote, R., Terro´n, J. and Go´mez, F. Monte Carlo correction factors for a Farmer 0.6 cm3 ion chamber dose measurement in the build-up region of the 6 MV clinical beam. Phys. Med. Biol. 51, 1523–1532 (2006). 10. Apipunyasopon, L., Srisatit, S. and Phaisangittisakul, N. An investigation of the depth dose in the build-up region, and surface dose for a 6-MV therapeutic photon beam: Monte Carlo simulation and measurements. J. Radiat. Res. 54, 374–382 (2012). 11. Nilsson, B. and Montelius, A. Fluence perturbation in photon beams under nonequilibrium conditions. Med. Phys. 13, 191 (1986).
12. Shea, E. O. and McCavana, P. Review of surface dose detectors in radiotherapy. J. Radiother. Pract. 3, 69–76 (2003). 13. Maniquis, V. Monte Carlo dose verification of an X-ray beam in a virtual water phantom. Georgia Institute of Technology (2006). 14. Rogers, D., Faddegon, B., Ding, G., Ma, C. M., We, J. and Mackie, T. BEAM: A Monte Carlo code to simulate radiotherapy treatment units. Med. Phys. 22, 503– 524 (1995). 15. Rogers, D. and Kawrakow, I. BEAMnrc users manual. NRC Report PIRS. National Research Council of Canada, pp. 1– 268 (2006). 16. Sheikh-Bagheri, D., Rogers, D. C., Ross, K. and Seuntjens, J. P. Comparison of measured and Monte Carlo calculated dose distributions from the NRC linac. Med. Phys. 27, 2256–2266 (2000). 17. Verhaegen, F. and Seuntjens, J. Monte Carlo modelling of external radiotherapy photon beams. Phys. Med. Biol. 48, 107–164 (2003). 18. Hill, R., Brown, S. and Baldock, C. Evaluation of the water equivalence of solid phantoms using gamma ray transmission measurements. Radiat. Meas. 43, 1258–1264 (2008). 19. Ramaseshan, R., Kohli, K., Cao, F. and Heaton, R. Dosimetric evaluation of Plastic Water DiagnosticTherapy. J. Appl. Clin. Med. Phys./Am. Coll. Med. Phys. 9, 2761– 2775 (2008). 20. Constantinou, C., Attix, F. and Paliwal, B. R. A solid water phantom material for radiotherapy x-ray and -ray beam calibrations. Med. Phys. 9, 436–441 (1982). 21. Almond, P. R., Biggs, P. J., Coursey, B., Hanson, W., Huq, M. S., Nath, R. and Rogers, D. AAPM’s TG-51 protocol for clinical reference dosimetry of high-energy photon and electron beams. Med. Phys. 26, 1847– 1870 (1999). 22. Andreo, P., Burns, D. T., Hohlfeld, K., Huq, M. S., Kanai, T., Laitano, F., Smyth, V. G. and Vynckier, S. Absorbed dose determination in external beam radiotherapy: An International Code of Practice for Dosimetry based on standards of absorbed dose to water. IAEA TRS-398. International Atomic Energy Agency 11b, pp. 13– 181 (2000). 23. Mesbahi, A., Reilly, A. J. and Thwaites, D. I. Development and commissioning of a Monte Carlo photon beam model for Varian Clinac 2100EX linear accelerator. Appl. Radiat. Isot. 64, 656–662 (2006). 24. Chan, K., Heng, S. M. and Smee, R. Application of Monte Carlo simulation in treatment planning for radiation oncology. InTech (2011). ISBN: 978-953-307-427-6. 25. Lin, J.-P., Chu, T.-C., Lin, S.-Y. and Liu, M.-T. Skin dose measurement by using ultra-thin TLDs. Appl. Radiat. Isot. 55, 383 –391 (2001).
doi:10.1093/rpd/nct315
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS, TLDS AND IONISATION CHAMBER AND COMPARISON WITH MONTE CARLO SIMULATION
*Corresponding author: [email protected] Received 12 September 2013; revised 31 October 2013; accepted 6 November 2013 Estimation of the surface dose is very important for patients undergoing radiation therapy. The purpose of this study is to investigate the dose at the surface of a water phantom at a depth of 0.007 cm as recommended by the International Commission on Radiological Protection and International Commission on Radiation Units and Measurement with radiochromic films (RFs), thermoluminescent dosemeters and an ionisation chamber in a 6-MV photon beam. The results were compared with the theoretical calculation using Monte Carlo (MC) simulation software (MCNP5, BEAMnrc and DOSXYZnrc). The RF was calibrated by placing the films at a depth of maximum dose (dmax) in a solid water phantom and exposing it to doses from 0 to 500 cGy. The films were scanned using a transmission high-resolution HP scanner. The optical density of the film was obtained from the red component of the RGB images using ImageJ software. The per cent surface dose (PSD) and percentage depth dose (PDD) curve were obtained by placing film pieces at the surface and at different depths in the solid water phantom. TLDs were placed at a depth of 10 cm in a solid water phantom for calibration. Then the TLDs were placed at different depths in the water phantom and were exposed to obtain the PDD. The obtained PSD and PDD values were compared with those obtained using a cylindrical ionisation chamber. The PSD was also determined using Monte Carlo simulation of a LINAC 6-MV photon beam. The extrapolation method was used to determine the PSD for all measurements. The PSD was 15.0+ + 3.6 % for RF. The TLD measurement of + 3.0 %. The theorthe PSD was 16.0+ + 5.0 %. The (0.6 cm3) cylindrical ionisation chamber measurement of the PSD was 50.0+ etical calculation using MCNP5 and DOSXYZnrc yielded a PSD of 15.0+ + 2.0 % and 15.7+ + 2.2 %. In this study, good agreement between PSD measurements was observed using RF and TLDs with the Monte Carlo calculation. However, the cylindrical chamber measurement yielded an overestimate of the PSD. This is probably due to the ionisation chamber calibration factor that is only valid in charged particle equilibrium condition, which is not achieved at the surface in the build-up region.
INTRODUCTION The absorbed dose at the surface of a patient or phantom irradiated with a beam of megavoltage X rays arises from electrons generated in air above the phantom, from electron backscatter within the phantom, and from electrons generated by any solid material in the beam. So, the skin-sparing effect for high-energy gamma- and X-ray photons may be reduced or even lost(1). The relevant dose specification depth depends on the biological effect considered. According to the International Commission on Radiological Protection(2) and International Commission on Radiation Units and Measurement(3) the skin dose should be measured at 0.007 cm, and this depth generally corresponds to the interface between the epidermis and dermis layers of the skin. The surface dose can be defined as the energy deposited within an infinitesimally small mass of tissue at the surface of the phantom(1). Avariety of dosemeters can be used, such as fixed-separation parallel-plate chambers, radiochromic films (RFs) and TLDs. However, there is no dosemeter that has an infinitesimally small sensitive volume, and the surface dose by
definition is inherently difficult to measure. Most treatment planning systems (TPS) fail to accurately predict the dose at the skin of the patient because the skin cannot be defined in the computer tomography slices accurately and also because the voxel size used by the TPS is much larger than the skin depth (0.007 cm)(4). But the effective point of measurements for RFs and TLDs are close to the skin depth. So, they can be used for investigating the per cent surface dose (PSD) defined as the central axis skin dose relative to maximum dose (Dmax) at a depth of 0.007 cm for a 10` 10 cm2 field size. In the build-up region, the charged particle equilibrium does not exist and therefore an accurate measurement of PSD using an ionisation chamber cannot be obtained. Therefore, the choice of the measurement device is of great importance in PSD measurements. The introduction of RFs has reduced some of the problems encountered with other conventional radiation dosemeters. The high spatial resolution of RFs makes them ideal for the measurement of dose distributions in regions of high-dose gradients in radiation fields. There are three recent new RF models the
# The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
Saleh Alashrah1,2, Sivamany Kandaiya2, Nabil Maalej3 and A. El-Taher1,4,* 1 Department of Physics, Qassim University, Qassim, Saudi Arabia 2 Universiti Sains Malaysia, Penang, Malaysia 3 Department of Physics, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia 4 Physics Department, Faculty of Science, Al-Azher University, Assuit, Egypt
S. ALASHRAH ET AL.
I0 OD ¼ Log I where I0 is the light intensity with no film present and I the light intensity after passing through the film. The OD is usually a non-linear response to dose and hence requires calibration. The film is normally calibrated using a calibrated ionisation chamber. This approach has been able to provide useful dosimetric information. The uncertainties in dose measurements based on any film dosimetry system results from the uncertainty due to the effective point of measurements, non-uniform thickness of the film’s sensitive layer, the densitometer used to measure the OD and non-uniform radiation field during exposure(7). Thermoluminescent dosemeters (TLD) are used clinically due to the following advantages: (i) wide useful dose range, (ii) small physical size, (iii) no need for high voltage or cables and (iv) tissue equivalence (LiF) for most radiation types(8). TLDs can be particularly used for the absorbed dose measurements in areas where dose measurement is difficult and not as part of a routine verification procedure. The error in the TLD reading was found to be within 5 %(8). Unlike RF the LiF TLD can be annealed and re-used many times. In the surface region, charged particle equilibrium does not exist, as in all transition zones between two different media. This will cause perturbation effects in ionisation chambers. A PTW 30013 0.6-cm3 ionisation chamber was used because of its high accuracy in the charged particle equilibrium region. The chambers of fixed volume will tend to indicate a larger dose
in the build-up region than the actual dose(9). This increase in the measured dose is mainly a result of secondary electrons scattered into the chamber volume from the walls of the chamber. For this reason, it is recommended to use an extrapolation chamber to estimate the PSD(10, 11). By measuring the ionisation per unit volume as a function of electrode spacing, the superficial dose can be estimated by extrapolating the ionisation curves to zero electrode spacing. The perturbation effects in the ionisation chamber used for build-up measurements have been studied(12). The fluence perturbation due to electrons emitted through the side walls that cause an overestimation of the surface dose, have been thoroughly investigated by measurements using film, extrapolation chambers and by calculations(12). The perturbation of the electron fluence in the build-up region is due to lack of equilibrium in the transport of different categories of electrons contributing to the ionisation in the chamber. The main contribution to the ionisation, especially for small plate separations, is due to electrons coming from the air and treatment head. These electrons may then be backscattered. Photons hitting the chamber may emit electrons in the front electrode, collector and through the side walls. These electrons may then be scattered into the chamber volume. Electrons hitting the phantom close to the chamber may be scattered into the chamber. A small contribution is also obtained from electrons produced by photon interactions in the chamber(12). In principle, Monte Carlo simulations have a high spatial resolution when the used voxel size is small. The Monte Carlo technique involves using known probability distributions that govern the physical interactions of photons and electrons in various materials to simulate random trajectories of individual particles. By keeping track of processes of interest for a large number of histories, information regarding the average quantities and their correlated distributions can be obtained as well as the statistical fluctuations of specific events. The use of MC methods in radiation therapy physics has increased over the past few decades as high-energy photon and electron beams are being used in radiotherapy. It is necessary to account for the electron transport for dosimetry and treatment planning systems(13). In 1995 the EGS4 code was released(14) and is now upgraded to EGSnrc (electron gamma shower)(15), which is used for simulating electron and photon beams. EGSnrc has been employed in the medical physics to study problems in radiation dosimetry and radiation therapy. In addition, it is used in diagnostic imaging, design and characterisation of medical devices and radiation protection. This code allows easy modelling of a series of linear accelerators (Siemens, Elekta and Varian linacs and NRC research linac) and has led to publications of many of papers dealing with photon and electron beams from accelerators(16, 17).
Page 2 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
XR-T, HS and EBT. They have been introduced by International Specialty Products (Wayne, NJ, USA). In this experiment, the RF model HS was used. It has a single sensitive layer designed for two-dimensional dose measurements in high-energy photon beams (.1 MeV)(1). RF uses radiochemical processes to impart a change in the optical absorbance of the film for specific wavelengths of light. It can be used to measure very high doses associated with industrial irradiators and accelerators. The films are affected by exposure to bright light, temperature and humidity. Furthermore, ultraviolet light produces an unwanted colouration of the film. Therefore, appropriate handling and storage conditions are required. The films should thus be stored in an opaque container and only taken out for experiments and readout (5). The optical density (OD) of the films after exposure is preserved for some time. The disadvantage of these films is that they can only be used once and should be scanned at a defined time after exposure preferably 2 d after the exposure(6). The reduction in light passing through the film is a measure of its ‘blackness’ or ‘optical density’ expressed by
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS
MATERIALS AND METHODS Solid water phantom (RW3)
RF measurements The HS GafChromicw film model (International Specialty Products, Wayne, NJD) has been developed as a more sensitive and uniform alternative to the GafChromicw MD-55 film. It was specifically designed for the measurement of absorbed dose in high-energy photon and electron beams (.1 MeV). According to the manufacturer, the HS model covers a dose range from 1 to 50 Gy. The HS film consists of a single active layer sandwiched between two sheets of clear, transparent polyester, each with a thickness of 97 mm and a density of 1.38 g cm23 (Figure 1). The HS active layer has a thickness of 40 mm. The films were placed at dmax in a solid water phantom (30` 30` 20 cm3) and in the centre of a 10` 10 cm2 field size of 6-MV photon beam (from a
Dfit ¼ a þ b netOD þ c netODn
ð1Þ
The third term in Equation (1) is introduced to account for the non-linear dose response in the highdose region close to the saturation level for a given film dosimetry system. In order to predict the error in measurements of an unknown dose while using the calibration curve for each dosimetry system, the expression for error propagation was used:
s2y ¼
X @y2 i
@x
:s2xi
ð2Þ
where sy is the total estimated uncertainty (standard deviation) for a dose determined and sxi (i ¼ 1, 2, 3) is the standard deviations for the net OD and the two fitting parameters b and c. From the last equations, it follows:
Figure 1. Structure and dimension of HS RF model used in this study.
sPDD
Page 3 of 7
ffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi s 2 s 2 max ¼ þ D Dmax
ð3Þ
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
However, water phantom measurements are difficult to set up and many detectors in radiation therapy are not waterproof(18). Solid water equivalent phantoms such as RMI457 solid water, plastic water, RW3 solid phantom (PTW Freiburg) and Perspex(19) are now available. The solid water phantom has similar properties such as physical density, relative electron density and effective atomic number as well as absorption and scattering of radiation to those of water(18). The electron density of the medium is important as the Compton scatter is a dominant process at highenergy photon beams (MV). Hill et al.(18) studied RMI457, plastic water, RW3 solid phantom and perspex using the source 99Tcm. The maximum difference in the absorbed dose results was 4 % between the phantoms and water. Slabs of the solid water phantom (RW3) were used in this study. The solid water phantom is an epoxy resin-based solid substitute of water. Its absorption and scattering properties is within 1.0 % of water. The physical density ranges from 1.039 to 1.049 g cm23. These characteristics allow the solid water phantom to be used in high energy (radiation therapy)) measurements(20). The dimension of the phantom is 30` 30 cm2 and the slabs thickness varied from 0.1 to 1 cm.
Varian 21 EX LINAC) as shown in Figure 1. The source-to-surface distance (SSD) was 100 cm. The films (3.5` 3 cm2) were exposed at different doses from 0 to 500 cGy for calibration. For each exposure, the film pieces were removed from their light-protecting envelope and irradiated. After 2 d from the exposure, an HP Scanjet 4890 scanner was used to scan the films. The scanner has a film holder for positioning the film pieces. Once the scanner is turned on, it is important to allow the scanner to warm-up and allows its temperature to stabilise. The exposed films were scanned six times and averaged to reduce the error in measurements. After the pieces of the films were scanned, the image J (http://rsb.info.nih.gov) program was used to extract the red component of the red– green –blue (RGB) image as the absorption spectrum of the RF exhibits a maximum response in the red region of the visible spectrum. Subsequently, the net OD of the irradiated film pieces was determined by subtracting the OD of an unexposed film. The relation between the doses and optical densities were plotted to obtain a calibration curve. Matlab 7.0.1 program was used to obtain the best fit curve of the calibration. In order to find the most suitable fit of the dose as a function of OD, the following criteria were used: (1) the fit function has to be monotonically increasing; (2) the fit function has to go through zero and finally; (3) chose the function that gives the minimum relative uncertainty for the fitting parameters. Based on these criteria, a family of fitting functions of the form was chosen:
S. ALASHRAH ET AL.
Ionisation chamber measurements The ionisation chamber used in the present work is a PTW 0.6-cm3 Farmer chamber (type 30013). The cavity length is 23 mm with the wall materials made of graphite and polymethyl methacrylate. Measurements with the ionisation chamber were performed in a water phantom (30` 30` 30 cm3) at different depths (0.5, 1, 1.5, . . . , 20 cm). All experiments were performed with a dose of 1 Gy at dmax, SSD ¼ 100 cm and 10` 10 cm2 field size at the surface of water phantom. The chamber was positioned with its axis perpendicular to the beam axis, as recommended by the manufacturer for reference dosimetry. The ionisation readings were taken at the effective point of ionisation chamber, i.e. the chamber was shifted towards the surface by 0.6` radius of the cavity to obtain the PDD curve as recommended by TG51(21) and TRS 398(22). Monte Carlo simulation
TLD Measurements ` 3.2 mm ` 0.89 Thirty-six TLD-100 chips (3.2 mm mm) were placed in a solid water phantom (30` 30` 20 cm3) at a depth of 10 cm using a TLD holder made of Perspex. The holder had holes of the same size as the TLDs to reduce the effect of air gaps. The TLDs were irradiated with a 6-MV photon beam at 10` 10 cm2 field size on the phantom surface with a 100-cm SSD. After the TLDs were exposed, they were read using a TLD reader (system 4000 HARSHAW). Each TLD has a characteristic calibration factor value which relates its response to dose. Subsequently, the surface dose was measured using the TLDs. The TLDs were handled with vacuum tweezers using a plastic nozzle to avoid scratching the surface of the chips. All experiments were performed at a dose of 1 Gy at d max.to the supra-linearity region of LiF(1). The effective point of measurement was chosen at the centre of the TLD chip. TLDs were placed at different depths (0, 1, 2, 3, 4 . . . , 190 mm) in the phantom and were exposed using 100 MU, 6-MV photon beam, 10` 10 cm2 field size at the phantom surface, and an SSD of 100 cm as shown in Figure 3.
MCNP5 and BEAMnrc were used to calculate the energy absorbed in the build-up and decay regions of the 6-MV photon beam irradiating perpendicularly to the water phantom. A photon beam model was simulated using the geometry and materials composition of a Varian 2100EX (Varian Medical Systems, Palo Alto, CA, USA). A schematic representation of the LINAC head and its components is shown in Figure 4. The LINAC head components, including the target, primary collimator, flattening filter and secondary collimator jaws, were simulated based on manufacturerprovided information. The SSD was simulated to be at 100 cm and the field size on the surface of water phantom to be 10` 10 cm2. In most calculations, a photon transport cut-off of 0.01 MeV and an electron kinetic energy transport cut-off of 0.10 MeV for MCNP5 were used but the cut-off energy for electron and photon was 0.7 and 0.01 MeV, respectively, for EGSnrc/BEAMnrc and EGSnrc/DOSXYZnrc. The energy of the electron source was 6 MeV and the Gaussian electron distribution was used in the simulation(23). For BEAMnrc, the ISOURC¼19 was used as a source with 6 MeV and the full width at half
Figure 2. The RF measurement set-up.
Figure 3. The TLD measurements set-up.
Page 4 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
where s and smax are the uncertainties in dose D at an depth d and Dmax at an depth dmax. The films were placed at different depths (0, 1, 2, 3, 4, 5, 7..., 190 mm) as shown in Figure 2. They were adjusted to be in the centre and perpendicular to the beam with a 10` 10 cm2 field size at the phantom surface. The SSD was 100 cm and the gantry angle was 08. The dose of 3 Gy was delivered at the depth of a maximum dose. For each exposure, the exposed film was removed and another film was placed in a new position. The HP4890 scanner was used for scanning the pieces of film and ImageJ program was used for estimating the average OD. The net OD were converted to dose by using the calibration curve and the dose was normalised to the maximum dose at dmax to obtain the PDD (PDD ¼ D/Dmax). Finally, the error in the dose was calculated.
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS
Figure 4. The geometry of the head of LINAC (2100 EX) (a) using BEAMnrc and (b) for MCNP5 simulation.
maximum (FWHM) of the distribution of the source in x and y directions was 0.15 cm. The number of particles history was 80` 106 for MCNP5 and it was 1` 109 for BEAMnrc and DOSXYZnrc. The detectors were simulated to be at the centre of the phantom and perpendicular to the beam. These detectors were cubic with dimensions 2` 2` 0.02 cm3 for MCNP5. They were distributed at different depths (1, 1.3, 1.5, 1.7, 1.9, 2, . . . 190 mm). The DOSXYZnrc was used to simulate the phantom (50` 50` 50 cm3). The detector size was 0.2` 0.2` 0.2 cm3 in the build-up region. All the software (MCNP5, BEAMnrc and DOSXYZnrc) ran on a window vista PC (Intel core TM 2 Quad Process, 2.6 GHz, 500 GB hard drive, 4 GB memory).
RESULTS AND DISCUSSION The calibration curve for the HS RF model is shown in Figure 5. The best fit curve of dose versus the measured net OD is determined using Equation (1). The best fit was obtained with n ¼ 5. Using MTLAB 7.0.1 the best values of the parameters b and c with the corresponding uncertainties were determined.
Figure 6. Per cent depth dose curves for a 6-MV photon beam at the build-up region using RF, TLD-100, ionisation chamber, Monte Carlo simulation using MCNP5 and EGSnrc/DOSXYZnrc.
To estimate the clinically relevant skin dose, the effective points of measurement for the dosemeters must be known. For the ionisation chamber, the effective point is at a distance equal to 0.6 r cavity from the geometric centre where r cavity is the radius of the cavity of the cylindrical ionisation. On the other hand, for RFs and TLDs, the effective point of measurement was defined at the midpoint of the sensitive layer. For Monte Carlo results, the effective point measurement was defined at the geometric centre of the detector. Figure 6 shows the PDD curve in the build-up region using the RF, TLD, ionisation chamber, DOSXYZnrc and MCNP5 using SSD ¼ 100 cm. The difference in the results is in the build-up region between TLDs and EGSnrc or MCNP5 was within +5 %. However, the difference between RF and Monte Carlo simulation was within 10 %. Other groups have reported similar differences between the
Page 5 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
Figure 5. Calibration curve for HS RF with best fitting (D ¼ b` net OD þ c` net OD5, where b ¼ 4.956+0.43 and c ¼ 2.198` 1028 + 1.38` 1028 and R 2 ¼ 0.9982).
S. ALASHRAH ET AL. Table 1. The difference in the results for RF, TLD-100 and MCNP5 used in this study for dose at a depth of 20 cm. Dosemeter or calculation method PTW 30013 Farmer ionization chamber HS radiochromic film TLD-100 (3.2` 3.2` 0.89) MCNP5 BEAMnrc
PDD (d ¼ 20 cm) (%) 38 38 40 40 40
CONCLUSION measured and calculated doses in the build-up region reaching up to 10 %(24). The Monte Carlo [MCNP5 or EGSnrc (BEAMnrc/DOSXYZnrc)] simulation with a small pixel size has a high resolution, and no electronic equilibrium is required and provides accurate results. Consequently, the Monte Carlo simulation has been used as a gold standard for dose calculation. The PTW 30013 Farmer ionization chamber readings agree with MCNP5 and DOSXYZnrc at a depth of .4 mm. For depths shallower than 4 mm, there is a large difference between the chamber reading and MC simulation results. The skin depth was assumed to be at the reference point of 70 mm; therefore, an extrapolation method needs to be used to estimate the percentage skin dose. According to HS RF, the percentage skin dose is 15.0+3.6 % and it is 16.0+5.0 % for TLD-100. These results agree with MCNP5 (15.0+2.0 %) and DOSXYZnrc (15.7+2.2 %) results. When the TLDs were used to measure the PSD, their results agreed with the published results(25). Devic et al.(1) reported a PSD measurement of 16 % using the Attix parallel plate ionisation chamber. On the other hand, the PSD measurement obtained in this study using cylindrical (0.6 cm3) ionisation chamber is 50.0+3.0 %. Therefore, for percentage skin measurements, a thin dosemeter has to be used. A cylindrical chamber reading gives an overestimate of the PSD due to the presence of partial charged particle equilibrium. Figure 7 shows the measured PDD curve in the region after dmax using RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5. There is a very good agreement between RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5 results. For instance, the PDD results at a depth of 20 cm are shown in Table 1. Based on the results of Figure 7 and Table 1, the maximum difference in the PDD at a depth of 20 cm is ,4 %. The data show that both RF and TLD gave a better estimate of PSD than that from the cylindrical ionisation chamber. A parallel-plate extrapolation chamber than a cylindrical chamber will give a more accurate estimate of PSD.
The skin dose was investigated at a depth of 0.007 cm experimentally and by Monte Carlo simulation. Experimental measurements were done using RFs, TLD-100 and a cylindrical ionisation chamber (0.6 cm3) in a solid water phantom. The PSD was 15.0+3.6 and 16.0+5.0 % for RF and TLD-100, respectively. Ionisation chamber measurements were extrapolated to the surface and yielded a PSD of 50.0+3.0 %. The Monte Carlo simulation does not require electronic equilibrium and has the advantage of high resolution due to small voxel size and provides accurate dose calculation. Consequently, the Monte Carlo simulation has been used as a gold standard for dosimetry. The PSDs using MCNP5, EGSnrc/ BEAMnrc and EGSnrc/DOSXYZnrc were 15.0+2.0 and 15.7+2.2 %. Therefore, there is a good agreement between RF and TLD measurement with Monte Carlo results. However, the PSD measurement using an ionisation chamber was 50.0+3.0 %, very different from other Monte Carlo results. The reason for the overestimate of dose measurement using the ionisation chamber in the build-up region is mainly due the contribution of secondary electrons scattered into the chamber volume. The measurements obtained in this study have also shown that there is an agreement between RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5 in the region after dmax. The maximum percentage difference between the dosemeters was ,4 % at a depth of 20 cm. Extrapolating the cylindrical chamber measurements to the surface yields an overestimate of the PSD. To obtain an accurate measurement of the skin dose which agrees with the Monte Carlo calculation, thin dosemeters should be used such as RF and TLD. The main advantage of RF is small size and high resolution. However, RF requires calibration and the OD response is not linear with dose. The TLD response is linear with dose but requires meticulous handling during calibration, irradiation and reading and are usually larger in thickness compared with RF. Both RF and TLD do not provide immediate reading and require post-irradiation processing.
Page 6 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
Figure 7. The PDD curves in the decay region using RF, TLD-100, ionisation chamber, DOSXYZnrc and MCNP5.
SKIN DOSE MEASUREMENTS USING RADIOCHROMIC FILMS
ACKNOWLEDGEMENTS The authors would like to thank Mr Hassan Al-Gamdi from Dhahran Medical Centre (Saudi ARAMCO) and to Dr Tarak from King Faisal Specialist Hospital and Research Centre for their help. REFERENCES
Page 7 of 7
Downloaded from http://rpd.oxfordjournals.org/ at National Dong Hwa University Library on April 10, 2014
1. Devic, S., Seuntjens, J., Abdel-Rahman, W., Evans, M., Olivares, M., Podgorsak, E., Vuong, T. and Soares, C. G. Accurate skin dose measurements using radiochromic film in clinical applications. Med. Phys. 33, 1116 (2006). 2. ICRP. Recommendations of the International Commission on Radiological Protection. ICRP Publication 60. ICRP 21(1–3). Pergamon (1991). 3. ICRU. Determination of dose equivalents resulting from external radiation sources. Report No. 39. International Commission on Radiation Units and Measurement (1985). 4. Stathakis, S., Paskalev, J., Li, K., Yang, J., Wang, L. and Ma, C. Ultra-thin TLDs for skin dose determination in high energy photon beams. Phys. Med. Biol. 51, 3549 (2006). 5. Butson, M. J., Yu, P. K. N., Cheung, T. and Metcalfe, P. Radiochromic film for medical radiation dosimetry. Mater. Sci. Eng. R Rep. 41, 61– 120 (2003). 6. Niroomand-Rad, A., Blackwell, C. R., Coursey, B. M., Gall, K. P., Galvin, J. M., McLaughlin, W. L., Meigooni, A. S., Nath, R., Rodgers, J. E. and Soares, C. G. Radiochromic film dosimetry: recommendations of AAPM radiation therapy committee task group 55. Med. Phys. 25, 2093 (1998). 7. Devic, S., Seuntjens, J., Sham, E., Podgorsak, E. B., Schmidtlein, C. R., Kirov, A. S. and Soares, C. G. Precise radiochromic film dosimetry using a flat-bed document scanner. Med. Phys. 32, 2245– 2253 (2005). 8. Aschan, C. Applicability of thermoluminescent dosimeters in x-ray organ dose determination and in the dosimetry of system and boron neutron capture radiotherapy. Helsingin Yliopiston Verkkojulkaisut (1999). 9. Pena, J., Sa´nchez-Doblado, F., Capote, R., Terro´n, J. and Go´mez, F. Monte Carlo correction factors for a Farmer 0.6 cm3 ion chamber dose measurement in the build-up region of the 6 MV clinical beam. Phys. Med. Biol. 51, 1523–1532 (2006). 10. Apipunyasopon, L., Srisatit, S. and Phaisangittisakul, N. An investigation of the depth dose in the build-up region, and surface dose for a 6-MV therapeutic photon beam: Monte Carlo simulation and measurements. J. Radiat. Res. 54, 374–382 (2012). 11. Nilsson, B. and Montelius, A. Fluence perturbation in photon beams under nonequilibrium conditions. Med. Phys. 13, 191 (1986).
12. Shea, E. O. and McCavana, P. Review of surface dose detectors in radiotherapy. J. Radiother. Pract. 3, 69–76 (2003). 13. Maniquis, V. Monte Carlo dose verification of an X-ray beam in a virtual water phantom. Georgia Institute of Technology (2006). 14. Rogers, D., Faddegon, B., Ding, G., Ma, C. M., We, J. and Mackie, T. BEAM: A Monte Carlo code to simulate radiotherapy treatment units. Med. Phys. 22, 503– 524 (1995). 15. Rogers, D. and Kawrakow, I. BEAMnrc users manual. NRC Report PIRS. National Research Council of Canada, pp. 1– 268 (2006). 16. Sheikh-Bagheri, D., Rogers, D. C., Ross, K. and Seuntjens, J. P. Comparison of measured and Monte Carlo calculated dose distributions from the NRC linac. Med. Phys. 27, 2256–2266 (2000). 17. Verhaegen, F. and Seuntjens, J. Monte Carlo modelling of external radiotherapy photon beams. Phys. Med. Biol. 48, 107–164 (2003). 18. Hill, R., Brown, S. and Baldock, C. Evaluation of the water equivalence of solid phantoms using gamma ray transmission measurements. Radiat. Meas. 43, 1258–1264 (2008). 19. Ramaseshan, R., Kohli, K., Cao, F. and Heaton, R. Dosimetric evaluation of Plastic Water DiagnosticTherapy. J. Appl. Clin. Med. Phys./Am. Coll. Med. Phys. 9, 2761– 2775 (2008). 20. Constantinou, C., Attix, F. and Paliwal, B. R. A solid water phantom material for radiotherapy x-ray and -ray beam calibrations. Med. Phys. 9, 436–441 (1982). 21. Almond, P. R., Biggs, P. J., Coursey, B., Hanson, W., Huq, M. S., Nath, R. and Rogers, D. AAPM’s TG-51 protocol for clinical reference dosimetry of high-energy photon and electron beams. Med. Phys. 26, 1847– 1870 (1999). 22. Andreo, P., Burns, D. T., Hohlfeld, K., Huq, M. S., Kanai, T., Laitano, F., Smyth, V. G. and Vynckier, S. Absorbed dose determination in external beam radiotherapy: An International Code of Practice for Dosimetry based on standards of absorbed dose to water. IAEA TRS-398. International Atomic Energy Agency 11b, pp. 13– 181 (2000). 23. Mesbahi, A., Reilly, A. J. and Thwaites, D. I. Development and commissioning of a Monte Carlo photon beam model for Varian Clinac 2100EX linear accelerator. Appl. Radiat. Isot. 64, 656–662 (2006). 24. Chan, K., Heng, S. M. and Smee, R. Application of Monte Carlo simulation in treatment planning for radiation oncology. InTech (2011). ISBN: 978-953-307-427-6. 25. Lin, J.-P., Chu, T.-C., Lin, S.-Y. and Liu, M.-T. Skin dose measurement by using ultra-thin TLDs. Appl. Radiat. Isot. 55, 383 –391 (2001).
