Technical note: characterization of spectral changes with measurement geometry and magnetic field strength in light guides used for scintillation dosimetry.

The purpose of this work was to characterize the stem-effect signal and the Cerenkov light ratio (CLR) in various light guides as functions of measurement geometry and magnetic field strength. Two PMMA-, two silica-, and one polystyrene-based light guides were considered in this work. Spectra measurements were performed as functions of depth, fiber-beam angle, and magnetic field strength using an optical spectrometer. All measurements were performed using a clinical linear accelerator at a nominal photon beam energy of 6 MV. Depths ranging from 1 cm to 10 cm, fiber-beam angles ranging from 90 degrees to 30 degrees, and magnetic field strengths ranging from 0 T to ± 1.40 T were investigated. The CLR was calculated from each spectrum by taking the ratio of the integral signal between 400 nm and 500 nm to the integral signal between 500 nm and 600 nm. A maximum increase of 80.5% in the stem-effect signal was observed in the magnetic field. Variations in spectral shape and, consequently, the CLR were observed for all of the fibers as functions of magnetic field strength and measurement geometry, particularly for wavelengths less than 400 nm. The plastic fibers exhibited decreases in the CLR as a function of magnetic field strength at all depths investigated, whereas the silica fibers exhibited increases in the CLR with decreasing magnetic field strength. A maximum variation of 11.1% in the CLR was observed for the polystyrene fiber due to the magnetic field. The sensitivity of the CLR to the magnetic field decreased as the fiber-beam angle decreased. The measured spectral response, shape, and CLR were found to be sensitive to the applied magnetic field strength and polarity where the variations in response were unique to each fiber.

Determination of consensus k Q values for megavoltage photon beams for the update of IAEA TRS-398.

The IAEA is currently coordinating a multi-year project to update the TRS-398 Code of Practice for the dosimetry of external beam radiotherapy based on standards of absorbed dose to water. One major aspect of the project is the determination of new beam quality correction factors, k Q , for megavoltage photon beams consistent with developments in radiotherapy dosimetry and technology since the publication of TRS-398 in 2000. Specifically, all values must be based on, or consistent with, the key data of ICRU Report 90. Data sets obtained from Monte Carlo (MC) calculations by advanced users and measurements at primary standards laboratories have been compiled for 23 cylindrical ionization chamber types, consisting of 725 MC-calculated and 179 experimental data points. These have been used to derive consensus k Q values as a function of the beam quality index TPR20,10 with a combined standard uncertainty of 0.6%. Mean values of MC-derived chamber-specific [Formula: see text] factors for cylindrical and plane-parallel chamber types in 60Co beams have also been obtained with an estimated uncertainty of 0.4%.

The determination of beam quality correction factors: Monte Carlo simulations and measurements.

Modern dosimetry protocols are based on the use of ionization chambers provided with a calibration factor in terms of absorbed dose to water. The basic formula to determine the absorbed dose at a user's beam contains the well-known beam quality correction factor that is required whenever the quality of radiation used at calibration differs from that of the user's radiation. The dosimetry protocols describe the whole ionization chamber calibration procedure and include tabulated beam quality correction factors which refer to 60Co gamma radiation used as calibration quality. They have been calculated for a series of ionization chambers and radiation qualities based on formulae, which are also described in the protocols. In the case of high-energy photon beams, the relative standard uncertainty of the beam quality correction factor is estimated to amount to 1%. In the present work, two alternative methods to determine beam quality correction factors are prescribed-Monte Carlo simulation using the EGSnrc system and an experimental method based on a comparison with a reference chamber. Both Monte Carlo calculations and ratio measurements were carried out for nine chambers at several radiation beams. Four chamber types are not included in the current dosimetry protocols. Beam quality corrections for the reference chamber at two beam qualities were also measured using a calorimeter at a PTB Primary Standards Dosimetry Laboratory. Good agreement between the Monte Carlo calculated (1% uncertainty) and measured (0.5% uncertainty) beam quality correction factors was obtained. Based on these results we propose that beam quality correction factors can be generated both by measurements and by the Monte Carlo simulations with an uncertainty at least comparable to that given in current dosimetry protocols.

Direct determination of k Q factors for cylindrical and plane-parallel ionization chambers in high-energy electron beams from 6 MeV to 20 MeV.

For the ionometric determination of the absorbed dose to water, D w, in high-energy electron beams from a clinical accelerator, beam quality dependent correction factors, k Q, are required. By using a water calorimeter, these factors can be determined experimentally and potentially with lower standard uncertainties than those of the calculated k Q factors, which are tabulated in various dosimetry protocols. However, one of the challenges of water calorimetry in electron beams is the small measurement depths in water, together with the steep dose gradients present especially at lower energies. In this investigation, water calorimetry was implemented in electron beams to determine k Q factors for different types of cylindrical and plane-parallel ionization chambers (NE2561, NE2571, FC65-G, TM34001) in 10 cm × 10 cm electron beams from 6 MeV to 20 MeV (corresponding beam quality index R 50 ranging from 1.9 cm to 7.5 cm). The measurements were carried out using the linear accelerator facility of the Physikalisch-Technische Bundesanstalt. Relative standard uncertainties for the k Q factors between 0.50% for the 20 MeV beam and 0.75% for the 6 MeV beam were achieved. For electron energies above 8 MeV, general agreement was found between the relative electron energy dependencies of the k Q factors measured and those derived from the AAPM TG-51 protocol and recent Monte Carlo-based studies, as well as those from other experimental investigations. However, towards lower energies, discrepancies of up to 2.0% occurred for the k Q factors of the TM34001 and the NE2571 chamber.

Spectral characterization of plastic scintillation detector response as a function of magnetic field strength.

The purpose of this work was to characterize intensity and spectral response changes in a plastic scintillation detector (PSD) as a function of magnetic field strength. Spectra measurements as a function of magnetic field strength were performed using an optical spectrometer. The response of both a PSD and PMMA fiber were investigated to isolate the changes in response from the scintillator and the noise signal as a function of magnetic field strength. All irradiations were performed in water at a photon beam energy of 6 MV. Magnetic field strengths of (0, ±0.35, ±0.70, ±1.05, and ±1.40) T were investigated. Four noise subtraction techniques were investigated to evaluate the impact on the resulting noise-subtracted scintillator response with magnetic field strength. The noise subtraction methods included direct spectral subtraction, the spectral method, and variants thereof. The PMMA fiber exhibited changes in response of up to 50% with magnetic field strength due to the directional light emission from [Formula: see text]erenkov radiation. The PSD showed increases in response of up to 10% when not corrected for the noise signal, which agrees with previous investigations of scintillator response in magnetic fields. Decreases in the [Formula: see text]erenkov light ratio with negative field strength were observed with a maximum change at -1.40 T of 3.2% compared to 0 T. The change in the noise-subtracted PSD response as a function of magnetic field strength varied with the noise subtraction technique used. Even after noise subtraction, the PSD exhibited changes in response of up to 5.5% over the four noise subtraction methods investigated.

Characterization of the radiation quality of 60Co therapy units by the fraction of air kerma attributable to scattered photons.

In this work we present a new parameter for characterizing the emitted photon spectra of (60)Co radiotherapy units. It is intended to propose this parameter for the revised DIN standard 6809-1. In the previous DIN regulation, it had been sufficient to state the nature of the radioactive material within the source. However, scatter processes within the radioactive material as well as the source housing and the collimator system influence the shape of the photon spectrum, with a noticeable contribution in the low-energy portion. The fraction of the air kerma for a given distance from the source, position and beam size in air comprising all contributions by scattered photons up to an upper energy limit for the emitted spectrum from (60)Co decay, will be proposed as a typical parameter. The new quantity, which is termed the 'fraction of air kerma attributable to scattered photons', P(E)(Scatter), has been calculated for E = 1.17 MeV and compared for four different Monte Carlo-simulated spectra of used (60)Co devices. Not included in this new formalism is the air kerma contribution by scattered photons in between the two lines of the (60)Co spectrum. A simple measurement procedure based on the signal ratio of two Farmer chamber detectors with different wall materials is discussed and its feasibility shown.

Experimental determination of pCo perturbation factors for plane-parallel chambers.

For plane-parallel chambers used in electron dosimetry, modern dosimetry protocols recommend a cross-calibration against a calibrated cylindrical chamber. The rationale for this is the unacceptably large (up to 3-4%) chamber-to-chamber variations of the perturbation factors (pwall)Co, which have been reported for plane-parallel chambers of a given type. In some recent publications, it was shown that this is no longer the case for modern plane-parallel chambers. The aims of the present study are to obtain reliable information about the variation of the perturbation factors for modern types of plane-parallel chambers, and-if this variation is found to be acceptably small-to determine type-specific mean values for these perturbation factors which can be used for absorbed dose measurements in electron beams using plane-parallel chambers. In an extensive multi-center study, the individual perturbation factors pCo (which are usually assumed to be equal to (pwall)Co) for a total of 35 plane-parallel chambers of the Roos type, 15 chambers of the Markus type and 12 chambers of the Advanced Markus type were determined. From a total of 188 cross-calibration measurements, variations of the pCo values for different chambers of the same type of at most 1.0%, 0.9% and 0.6% were found for the chambers of the Roos, Markus and Advanced Markus types, respectively. The mean pCo values obtained from all measurements are [Formula: see text] and [Formula: see text]; the relative experimental standard deviation of the individual pCo values is less than 0.24% for all chamber types; the relative standard uncertainty of the mean pCo values is 1.1%.

Uncertainties in Monte Carlo-based absorbed dose calculations for an experimental benchmark.

There is a need to verify the accuracy of general purpose Monte Carlo codes like EGSnrc, which are commonly employed for investigations of dosimetric problems in radiation therapy. A number of experimental benchmarks have been published to compare calculated values of absorbed dose to experimentally determined values. However, there is a lack of absolute benchmarks, i.e. benchmarks without involved normalization which may cause some quantities to be cancelled. Therefore, at the Physikalisch-Technische Bundesanstalt a benchmark experiment was performed, which aimed at the absolute verification of radiation transport calculations for dosimetry in radiation therapy. A thimble-type ionization chamber in a solid phantom was irradiated by high-energy bremsstrahlung and the mean absorbed dose in the sensitive volume was measured per incident electron of the target. The characteristics of the accelerator and experimental setup were precisely determined and the results of a corresponding Monte Carlo simulation with EGSnrc are presented within this study. For a meaningful comparison, an analysis of the uncertainty of the Monte Carlo simulation is necessary. In this study uncertainties with regard to the simulation geometry, the radiation source, transport options of the Monte Carlo code and specific interaction cross sections are investigated, applying the general methodology of the Guide to the expression of uncertainty in measurement. Besides studying the general influence of changes in transport options of the EGSnrc code, uncertainties are analyzed by estimating the sensitivity coefficients of various input quantities in a first step. Secondly, standard uncertainties are assigned to each quantity which are known from the experiment, e.g. uncertainties for geometric dimensions. Data for more fundamental quantities such as photon cross sections and the I-value of electron stopping powers are taken from literature. The significant uncertainty contributions are identified as the energy of the radiation source and the underlying photon cross sections as well as the I-value of media involved in the simulation. The combined standard uncertainty of the Monte Carlo calculation yields 0.78% as a conservative estimation. The result of the calculation is close to the experimental result and with each combined standard uncertainty <1%, the accuracy of EGSnrc is confirmed. The setup and methodology of this study can be employed to benchmark other Monte Carlo codes for the calculation of absorbed dose in radiotherapy.

Experimental determination of kQ factors for cylindrical ionization chambers in 10 cm × 10 cm and 3 cm × 3 cm photon beams from 4 MV to 25 MV.

For the ionometric determination of absorbed dose to water, Dw, in megavoltage photon beams from a linear accelerator, beam-quality-dependent correction factors, kQ, are used for the ionization chambers. By using a water calorimeter, these factors can be determined experimentally and with substantially lower standard uncertainties compared to calculated values of the kQ, which are published in various dosimetry protocols. In this investigation, kQ for different types of cylindrical ionization chambers (NE 2561, NE 2571, FC 65 G) were determined experimentally in 10 cm × 10 cm photon beams from 4 MV to 25 MV (corresponding beam quality index TPR20,10 from 0.64 to 0.80). The measurements were carried out at the linear accelerator facility of the Physikalisch-Technische Bundesanstalt. It is shown that the kQ factors for a single ionization chamber in 10 cm × 10 cm photon beams can be measured with a relative standard uncertainty of 0.31%. In addition to these measurements in 10 cm × 10 cm fields, kQ factors for the NE 2561 chamber were also determined in smaller 3 cm × 3 cm photon beams between 6 MV and 25 MV. In this case, relative standard uncertainties between 0.35 % and 0.38 % are achieved for the kQ factors. It is found for this ionization chamber, that the ratio of the kQ factors in 3 cm × 3 cm and in 10 cm × 10 cm beams increases with increasing TPR20,10 to reach a value of 1.0095 at TPR20,10 = 0.8 with a relative standard uncertainty of 0.4 %.

Difference in the relative response of the alanine dosimeter to megavoltage x-ray and electron beams.

In order to increase the usefulness of the alanine dosimeter as a tool for quality assurance measurements in radiotherapy using MV x-rays, the response with respect to the dose to water needs to be known accurately. This quantity is determined experimentally relative to (60)Co for 4, 6, 8, 10, 15 and 25 MV x-rays from two clinical accelerators. For the calibration, kQ factors for ionization chambers with an uncertainty of 0.31% obtained from calorimetric measurements were used. The results, although not inconsistent with a constant difference in response for all MV x-ray qualities compared to (60)Co, suggest a slow decrease from approximately 0.996 at low energies (4-6 MV) to 0.989 at the highest energy, 25 MV. The relative uncertainty achieved for the relative response varies between 0.35% and 0.41%. The results are confirmed by revised experimental data from the NRC as well as by Monte Carlo simulations using a density correction for crystalline alanine. By comparison with simulated and measured data, also for MeV electrons, it is demonstrated that the weak energy dependence can be explained by a transition of the alanine dosimeter (with increasing MV values) from a photon detector to an electron detector. An in-depth description of the calculation of the results and the corresponding uncertainty components is presented in an appendix for the interested reader. With respect to previous publications, the uncertainty budget had to be modified due to new evidence and to changes of the measurement and analysis method used at PTB for alanine/ESR.

RBE of nearly monoenergetic neutrons at energies of 36 keV-14.6 MeV for induction of dicentrics in human lymphocytes.

We examined the induction of dicentric chromosomes in human lymphocytes irradiated in vitro with nearly monoenergetic neutrons at energies in the range of 36 keV-15.0 MeV. For the assessment of the relative biological effectiveness (RBE) both 220 kV x-rays and (60)Co gamma-rays were used as reference radiations. To avoid potential confounding factors that would influence the outcome of the experiments, only blood from one individual was used. The neutron RBE culture conditions ensured that the chromosome analysis could be performed exclusively in metaphases of the first cell cycle in vitro. For the reference radiation of 220 kV x-rays, the values of RBE(M) were found to increase from 16.6 (E(n)=36 keV) to the maximum value of 23.4 (E(n)=385 keV). For (60)Co gamma-rays utilized as the reference radiation, the corresponding RBE(M) values were found to be higher by a factor of 4. These results agree well with the previously published large data sets of three laboratories on dose-response relationships for dicentrics or dicentrics plus centric rings. They show a similar dependence of RBE on neutron energy.