Grid/lattice therapy: consideration of small field dosimetry.

Small-field dosimetry used in special procedures such as gamma knife, Cyberknife, Tomotherapy, IMRT, and VMAT has been in evolution after several radiation incidences with very significant (70%) errors due to poor understanding of the dosimetry. IAEA-TRS-483 and AAPM-TG-155 have provided comprehensive information on small-fields dosimetry in terms of code of practice and relative dosimetry. Data for various detectors and conditions have been elaborated. It turns out that with a suitable detectors dose measurement accuracy can be reasonably (±3%) achieved for 6 MV beams for fields >1×1 cm2. For grid therapy, even though the treatment is performed with small fields created by either customized blocks, multileaf collimator (MLC), or specialized devices, it is multiple small fields that creates combined treatment. Hence understanding the dosimetry in collection of holes of small field is a separate challenge that needs to be addressed. It is more critical to understand the scattering conditions from multiple holes that form the treatment grid fields. Scattering changes the beam energy (softer) and hence dosimetry protocol needs to be properly examined for having suitable dosimetric parameters. In lieu of beam parameter unavailability in physical grid devices, MLC-based forward and inverse planning is an alternative path for bulky tumours. Selection of detectors in small field measurement is critical and it is more critical in mixed beams created by scattering condition. Ramification of small field concept used in grid therapy along with major consideration of scattering condition is explored. Even though this review article is focussed mainly for dosimetry for low-energy megavoltage photon beam (6 MV) but similar procedures could be adopted for high energy beams. To eliminate small field issues, lattice therapy with the help of MLC is a preferrable choice.

Dependence of surface dose on atomic number in megavoltage photon beams.

BACKGROUND: Surface dose in megavoltage photon radiotherapy has a significant clinical impact on the skin-sparing effect. In previously published works, it was established that the presence of medium atomic number (Z) absorbers, such as tin, decreases the surface dose. It was concluded that high-Z absorbers, such as lead, increase the surface dose, relative to medium-Z absorbers, due to the increased contributions from photoelectrons and electron-positron pairs. PURPOSE: The purpose of this investigation is to revisit these conclusions in the context of photon beams from modern linacs. METHODS: A metric estimating the relative intensity of charged particles emitted in the forward direction, I f ${I}_f$ , was proposed using cross-sections for the photon interactions. The I f ${I}_f$ values were calculated for various absorbers using energy spectra of 6 and 10 MV photon beams from a Varian TrueBeam linac. Monte Carlo (MC) simulations were performed using TOPAS MC code to calculate the surface dose for various absorbers. Surface dose measurements were performed with 6 and 10 MV photon beams with tin and lead absorbers. RESULTS: The I f ${I}_f$ values were found to decrease as a function of Z for both 6 and 10 MV photon beams indicating that the surface dose from electrons emitted in the forward direction consistently decreases with increasing Z. With the increasing Z of the absorbers, both experimental and MC-calculated surface dose decreased without exhibiting a minimum at medium-Z absorbers. The surface dose for lead and tin was determined to be within 1% of each other for both 6 and 10 MV photon beams using MC simulations and experimental measurements. Therefore, no statistically significant difference in surface dose was found between the tin and lead absorbers disproving the presence of any minima in the surface dose versus the Z curve as has been reported in the literature. CONCLUSIONS: Surface dose for modern photon beams can be reduced using both medium and high Z absorbers since a consistent decrease in surface dose was found with increasing absorber Z.

Characterization of a segmented printed circuit board (PCB) as a standard for absorbed dose to water from alpha-emitting radionuclides.

BACKGROUND: Our previous work introduced and evaluated a standard for surface absorbed dose rate per unit radioactivity to water from unsealed alpha-emitting radionuclides used in targeted radionuclide therapy (TRT). An overall uncertainty over 4.0% at k = 1 was reported for the absorbed dose to air measurements, which was partially attributed to the rotational alignment uncertainty in the geometrical setup. PURPOSE: A printed circuit board (PCB) with a segmented guard was constructed to align the extrapolation chamber (EC) and the source plates using a differential capacitance technique. The PCB EC aimed to enhance the repeatability of the ionization current measurements. The PCB EC was evaluated using a thin film 210Po source. The measured absorbed dose to air cavity was compared with the Monte Carlo (MC) calculations. Using the extrapolation method, the surface absorbed dose rate to water was calculated. METHODS: The PCB EC was constructed with a 4.50 mm diameter collector surrounded by four sectors and a guard electrode. The sectors were isolated for rotational alignment and later connected to the guard for ionization current measurements. A bridge circuit measured differential capacitance between opposing sectors, and a hexapod motion stage rotated the source substrate to minimize the differential capacitance. The EC was evaluated using a 210Po source with a 3.20 mm diameter and 1.253 µ $\mu $ Ci radioactivity. MC simulations were performed to calculate the k p o i n t ${k}_{point}$ , k b a c k s c a t t e r ${k}_{backscatter}$ , and k d i v ${k}_{div}$ correction factors. Ionization current measurements were performed for air gaps in the 0.3-0.525 mm range and surface absorbed dose rate to water was calculated. RESULTS: Rotational offsets of up to 3.0° were found and the current repeatability was found to increase with the absorbed dose to air uncertainty calculated to be ∼2.0%. Using the capacitance method, the effective EC diameter was measured to be 4.53 mm. The recombination, polarity, and electrometer corrections were reported to be within 1.00% across all measurement trials. The MC-calculated correction factors were calculated to be much larger than the recombination and polarity correction factors. The average k p o i n t ${k}_{point}$ , k b a c k s c a t t e r ${k}_{backscatter}$ , and k d i v ${k}_{div}$ corrections were calculated to be 1.063, 0.9402, and 2.136, respectively. The MC-calculated absorbed dose to air was found to overestimate the absorbed dose by over 4.00% when compared with the measured absorbed dose to air. The surface absorbed dose rate to water was calculated to be 2.304 × 10 - 6 $2.304 \times {10}^{ - 6}$ Gy/s/Bq with an overall uncertainty of 4.07%. CONCLUSIONS: The constructed PCB EC was deemed suitable as an absorbed dose standard. A repeatable rotational alignment was achieved using the differential capacitance technique. The metal electrodes on the PCB made a difference of < 1.00% on the backscatter correction when compared to the EC comprised of polystyrene-equivalent collector. A 20% difference in the surface absorbed dose rate to water was found between the two ECs, which is attributed to the cavity diameter differences leading to different magnitudes of dose fall-off along the lateral direction.

A diffusion-leakage model coupled with dose point kernels (DPK) for dosimetry of diffusing alpha-emitters radiation therapy (DaRT).

BACKGROUND: Diffusing alpha-emitters radiation therapy (DaRT) is a novel brachytherapy technique that leverages the diffusive flow of 224Ra progeny within the tumor volume over the course of the treatment. Cell killing is achieved by the emitted alpha particles that have a short range in tissue and high linear energy transfer. The current proposed absorbed dose calculation method for DaRT is based on a diffusion-leakage (DL) model that neglects absorbed dose from beta particles. PURPOSE: This work aimed to couple the DL model with dose point kernels (DPKs) to account for dose from beta particles as well as to consider the non-local deposition of energy. METHODS: The DaRT seed was modeled using COMSOL multiphysics and the DL model was implemented to extract the spatial information of the diffusing daughters. Using Monte-Carlo (MC) methods, DPKs were generated for 212Pb, 212Bi, and their progenies since they were considered to be the dominant beta emitters in the 224Ra radioactive decay chain. A convolution operation was performed between the integrated number densities of the diffusing daughters and DPKs to calculate the total absorbed dose over a 30-day treatment period. Both high-diffusion and low-diffusion cases were considered. RESULTS: The calculated DPKs showed non-negligible energy deposition over several millimeters from the source location. An absorbed dose >10 Gy was deposited within a 1.8 mm radial distance for the low diffusion case and a 2.2 mm radial distance for the high diffusion case. When the DPK method was compared with the local energy deposition method that solely considered dose from alpha particles, differences above 1 Gy were found within 1.3 and 1.8 mm radial distances from the surface of the source for the low diffusion and high diffusion cases, respectively. CONCLUSIONS: The proposed method enhances the accuracy of the dose calculation method used for the DaRT technique.

Dose-rate dependence and IMRT QA suitability of EBT3 radiochromic films for pulse reduced dose-rate radiotherapy (PRDR) dosimetry.

BACKGROUND: Pulsed reduced dose rate (PRDR) is an emerging radiotherapy technique for recurrent diseases. It is pertinent that the linac beam characteristics are evaluated for PRDR dose rates and a suitable dosimeter is employed for IMRT QA. PURPOSE: This study sought to investigate the pulse characteristics of a 6 MV photon beam during PRDR irradiations on a commercial linac. The feasibility of using EBT3 radiochromic film for use in IMRT QA was also investigated by comparing its response to a commercial diode array phantom. METHODS: A plastic scintillator detector was employed to measure the photon pulse characteristics across nominal repetition rates (NRRs) in the 5-600 MU/min range. Film was irradiated with dose rates in the 0.033-4 Gy/min range to study the dose rate dependence. Five clinical PRDR treatment plans were selected for IMRT QA with the Delta4 phantom and EBT3 film sheets. The planned and measured dose were compared using gamma analysis with a criterion of 3%/3 mm. EBT3 film QA was performed using a cumulative technique and a weighting factor technique. RESULTS: Negligible differences were observed in the pulse width and height data between the investigated NRRs. The pulse width was measured to be 3.15 ± 0.01 µ s $\mu s$ and the PRF was calculated to be 3-357 Hz for the 5-600 MU/min NRRs. The EBT3 film was found to be dose rate independent within 3%. The gamma pass rates (GPRs) were above 99% and 90% for the Delta4 phantom and the EBT3 film using the cumulative QA method, respectively. GPRs as low as 80% were noted for the weighting factor EBT3 QA method. CONCLUSIONS: Altering the NRRs changes the mean dose rate while the instantaneous dose rate remains constant. The EBT3 film was found to be suitable for PRDR dosimetry and IMRT QA with minimal dose rate dependence.

An independent Monte Carlo-based IMRT QA tool for a 0.35 T MRI-guided linear accelerator.

PURPOSE: To develop an independent log file-based intensity-modulated radiation therapy (IMRT) quality assurance (QA) tool for the 0.35 T magnetic resonance-linac (MR-linac) and investigate the ability of various IMRT plan complexity metrics to predict the QA results. Complexity metrics related to tissue heterogeneity were also introduced. METHODS: The tool for particle simulation (TOPAS) Monte Carlo code was utilized with a previously validated linac head model. A cohort of 29 treatment plans was selected for IMRT QA using the developed QA tool and the vendor-supplied adaptive QA (AQA) tool. For 27 independent patient cases, various IMRT plan complexity metrics were calculated to assess the deliverability of these plans. A correlation between the gamma pass rates (GPRs) from the AQA results and calculated IMRT complexity metrics was determined using the Pearson correlation coefficients. Tissue heterogeneity complexity metrics were calculated based on the gradient of the Hounsfield units. RESULTS: The median and interquartile range for the TOPAS GPRs (3%/3 mm criteria) were 97.24% and 3.75%, respectively, and were 99.54% and 0.36% for the AQA tool, respectively. The computational time for TOPAS ranged from 4 to 8 h to achieve a statistical uncertainty of <1.5%, whereas the AQA tool had an average calculation time of a few minutes. Of the 23 calculated IMRT plan complexity metrics, the AQA GPRs had correlations with 7 out of 23 of the calculated metrics. Strong correlations (|r| > 0.7) were found between the GPRs and the heterogeneity complexity metrics introduced in this work. CONCLUSIONS: An independent MC and log file-based IMRT QA tool was successfully developed and can be clinically deployed for offline QA. The complexity metrics will supplement QA reports and provide information regarding plan complexity.

On the perturbation effect and LET dependence of beam quality correction factors in carbon ion beams.

BACKGROUND: In a recent study, we reported beam quality correction factors, fQ , in carbon ion beams using Monte Carlo (MC) methods for a cylindrical and a parallel-plate ionization chamber (IC). A non-negligible perturbation effect was observed; however, the magnitude of the perturbation correction due to the specific IC subcomponents was not included. Furthermore, the stopping power data presented in the International Commission on Radiation Units and Measurements (ICRU) report 73 were used, whereas the latest stopping power data have been reported in the ICRU report 90. PURPOSE: The aim of this study was to extend our previous work by computing fQ correction factors using the ICRU 90 stopping power data and by reporting IC-specific perturbation correction factors. Possible energy or linear energy transfer (LET) dependence of the fQ correction factor was investigated by simulating both pristine beams and spread-out Bragg peaks (SOBPs). METHODS: The TOol for PArticle Simulation (TOPAS)/GEANT4 MC code was used in this study. A 30 × 30 × 50 cm3 water phantom was simulated with a uniform 10 × 10 cm2 parallel beam incident on the surface. A Farmer-type cylindrical IC (Exradin A12) and two parallel-plate ICs (Exradin P11 and A11) were simulated in TOPAS using the manufacturer-provided geometrical drawings. The fQ correction factor was calculated in pristine carbon ion beams in the 150-450 MeV/u energy range at 2 cm depth and in the middle of the flat region of four SOBPs. The kQ correction factor was calculated by simulating the fQo correction factor in a 60 Co beam at 5 cm depth. The perturbation correction factors due to the presence of the individual IC subcomponents, such as the displacement effect in the air cavity, collecting electrode, chamber wall, and chamber stem, were calculated at 2 cm depth for monoenergetic beams only. Additionally, the mean dose-averaged and track-averaged LET was calculated at the depths at which the fQ was calculated. RESULTS: The ICRU 90 fQ correction factors were reported. The pdis correction factor was found to be significant for the cylindrical IC with magnitudes up to 1.70%. The individual perturbation corrections for the parallel-plate ICs were <1.0% except for the A11 pcel correction at the lowest energy. The fQ correction for the P11 IC exhibited an energy dependence of >1.00% and displayed differences up to 0.87% between pristine beams and SOBPs. Conversely, the fQ for A11 and A12 displayed a minimal energy dependence of <0.50%. The energy dependence was found to manifest in the LET dependence for the P11 IC. A statistically significant LET dependence was found only for the P11 IC in pristine beams only with a magnitude of <1.10%. CONCLUSIONS: The perturbation and kQ correction factor should be calculated for the specific IC to be used in carbon ion beam reference dosimetry as a function of beam quality.

A multi-institutional comparison of acceptance testing data for a 0.35 T MRI scanner.

Objective.To present and quantify the variability in the acceptance testing data for the imaging component of the 0.35 T magnetic resonance-linear accelerator (MR-linac).Approach.The current acceptance testing protocol by the MR-linac vendor was described along with the equipment and scanner parameters utilized throughout the process. TheBofield homogeneity, SNR/uniformity of the combined and individual receiver coils, American College of Radiology (ACR) image quality testing, and spatial integrity of the imaging data were collected from twelve different institutions. The variability in the results was accentuated and the ramifications of the results were discussed in the context of MR-guided radiation therapy.Main Results.TheBofield homogeneity was found to have a large gantry dependence with the median values being <4 ppm for all gantry angles. The SNR and uniformity were found to be well above the vendor-specified thresholds with a relatively small institutional-dependence. All institutions passed the ACR image uniformity tests. The largest institutional variability was noted to be for the slice positional accuracy test. The spatial fidelity was calculated to be <1.0 and <2.1 mm within a 100 and a 175 mm radius from the isocenter.Significance.The results from this study can be used to set the tolerances and formal guidelines for MR-linacs imaging quality assurance. Additionally, the multi-institutional data reported in this work will aid in future MR-linac acceptance and commissioning.

On the length used for CT ionization chambers to determine CTDI.

Objective.Computed tomography dose index (CTDI) calculations based on measurements made with CT ionization chambers require characterization of two chamber properties: radiation sensitivity and effective length. The sensitivity of a CT ionization chamber is currently determined in some countries by calibration in an x-ray field that irradiates the entire chamber. Determination of the effective length is left to the user, and this value is frequently assumed to be equivalent to the nominal length-typically 100 mm-stated by the manufacturer. This assumption undermines the intention and usefulness of CTDI calculation. Thus, a slit-based calibration,NKL, of the CT ionization chambers was proposed by collimating the x-ray beam to a well-defined aperture width. The aim of this work is to compare the two methods.Approach.Four different CT ionization chambers (Standard Imaging Exradin A101, Radcal 10x5-3CT, Victoreen 500-100, and Capintec PC-4P) are investigated in this work. Sensitivity profiles were measured for all four chambers and effective/rated chamber lengths were calculated. A novel Monte-Carlo based correction was proposed to account for the presence of the aperture. CTDI was calculated and compared for two calibration beams as well as for a commercial CT scanner using Exradin A101 and Radcal 10x5-3CT chambers.Main results.The nominal chamber length was found to deviate up to 21% compared to the effective length. Correction for the aperture depended on the aperture opening size. CTDI calculation results illustrate the potential 17% error in CTDI calculation that can be caused by assuming the effective chamber length is equivalent to the manufacturer's stated nominal length. CTDI calculations with CT ionization chambers calibrated with an air-kerma length calibration method yield the smallest variation in the CTDI regardless of the chamber model.Significance.To avoid an erroneous CTDI, information regarding the chamber's effective length must be included in the calibration or stated by the manufacturer. Alternatively, a slit-based calibration can be performed.

A multi-institutional comparison of dosimetric data for a 0.35 T MR-linac.

Objective.A comparison of percent depth dose (PDD) curves, lateral beam profiles, output factors (OFs), multileaf collimator (MLC) leakage, and couch transmission factors was performed between ten institutes for a commercial 0.35 T MR-linac.Approach.The measured data was collected during acceptance testing of the MR-linac. The PDD curves were measured for the 3.32 × 3.32 cm2, 9.96 × 9.96 cm2, and 27.20 × 24.07 cm2field sizes. The lateral beam profiles were acquired for a 27.20 × 24.07 cm2field size using an ion chamber array and penumbra was defined as the distance between 80% of the maximum dose and 20% of the maximum dose after normalizing the profiles to the dose at the inflection points. The OFs were measured using solid-state dosimeters, whereas radiochromic films were utilized to measure radiation leakage through the MLC stacks. The relative couch transmission factors were measured for various gantry angles. The variation in the multi-institutional data was quantified using the percent standard deviation metric.Main results.Minimal variations (<1%) were found between the PDD data, except for the build-up region and the deeper regions of the PDD curve. The in-field region of the lateral beam profiles varied <1.5% between different institutions and a small variation (<0.7 mm) in penumbra was observed. A variation of <1% was observed in the OF data for field sizes above 1.66 × 1.66 cm2, whereas large variations were shown for small-field sizes. The average and maximum MLC leakage was calculated to be <0.3% and <0.6%, which was well below the international electrotechnical commission (IEC) leakage thresholds. The couch transmission was smallest for oblique beams and ranged from 0.83 to 0.87.Significance.The variation in the data was found to be relatively small and the different 0.35 T MR-linacs were concluded to have similar dosimetric characteristics.

Monte Carlo-derived ionization chamber correction factors in therapeutic carbon ion beams.

The accuracy of electromagnetic transport in the GEANT4 Monte Carlo (MC) code was investigated for carbon ion beams and ionization chamber (IC)-specific beam quality correction factors were calculated. This work implemented a Fano cavity test for carbon ion beams in the 100-450 MeV/u energy range to assess the accuracy of the default electromagnetic physics parameters. TheUrbanand theWentzel-VImultiple Coulomb scattering models were evaluated and the impact ofmaxStep,dRover,andfinal rangeparameters on the accuracy of the transport algorithm was investigated. The optimal production thresholds for an accurate calculation offQvalues, which is the product of the water-to-air stopping power ratio and the IC-specific perturbation correction factor, were also studied. ThefQcorrection factors were calculated for a cylindrical and a parallel-plate IC using carbon ions in the 150-450 MeV/u energy range. Modifying the default electromagnetic physics parameters resulted in a maximum deviation from theory of 0.3%. Therefore, the default EM parameters were used for the remainder of this work. ThefQfactors were found to converge for both ICs with decreasing production threshold distance below 5µm. ThefQvalues obtained in this work agreed with the TRS-398 stopping power ratios and other previously reported results within uncertainty. This study highlights an accurate MC-based technique to calculate the combined stopping power ratio and the perturbation correction factor for any IC in carbon ion beams.

Evaluation of the GEANT4 transport algorithm and radioactive decay data for alpha particle dosimetry.

A Fano cavity test was implemented in GEANT4 Monte Carlo code to evaluate the alpha particle transport algorithm. GEANT4 alpha emission data for 212Pb, 223Ra, 227Th, and 225Ac was compared with the MIRD and RADAR decay databases. Optimal electromagnetic transport parameters (dRover of 0.1 and final range of 1 µm) were recommended since the calculated results with the default parameters differed up to 4.7% from the theoretical results. Good agreement was found between the three decay databases besides a few discrepancies.

Development and evaluation of a GEANT4-based Monte Carlo Model of a 0.35 T MR-guided radiation therapy (MRgRT) linear accelerator.

PURPOSE: The aim of this work was to develop and benchmark a magnetic resonance (MR)-guided linear accelerator head model using the GEANT4 Monte Carlo (MC) code. The validated model was compared to the treatment planning system (TPS) and was also used to quantify the electron return effect (ERE) at a lung-water interface. METHODS: The average energy, including the spread in the energy distribution, and the radial intensity distribution of the incident electron beam were iteratively optimized in order to match the simulated beam profiles and percent depth dose (PDD) data to measured data. The GEANT4 MC model was then compared to the TPS model using several photon beam tests including oblique beams, an off-axis aperture, and heterogeneous phantoms. The benchmarked MC model was utilized to compute output factors (OFs) with the 0.35 T magnetic field turned on and off. The ERE was quantified at a lung-water interface by simulating PDD curves with and without the magnetic field for 6.6 × 6.6  cm 2 and 2.5 × 2.5  cm 2 field sizes. A 2%/2 mm gamma criterion was used to compare the MC model with the TPS data throughout this study. RESULTS: The final incident electron beam parameters were 6.0 MeV average energy with a 1.5 MeV full width at half maximum (FWHM) Gaussian energy spread and a 1.0 mm FWHM Gaussian radial intensity distribution. The MC-simulated OFs were found to be in agreement with the TPS-calculated and measured OFs, and no statistical difference was observed between the 0.35 T and 0.0 T OFs. Good agreement was observed between the TPS-calculated and MC-simulated data for the photon beam tests with gamma pass rates ranging from 96% to 100%. An increase of 4.3% in the ERE was observed for the 6.6 × 6.6  cm 2 field size relative to the 2.5 × 2.5  cm 2 field size. The ratio of the 0.35 T PDD to the 0.0 T PDD was found to be up to 1.098 near lung-water interfaces for the 6.6 × 6.6  cm 2 field size using the MC model. CONCLUSIONS: A vendor-independent Monte Carlo model has been developed and benchmarked for a 0.35 T/6 MV MR-linac. Good agreement was obtained between the GEANT4 and TPS models except near heterogeneity interfaces.

A Monte Carlo Investigation of Dose Point Kernel Scaling for α-Emitting Radionuclides.

Background: Density-based dose point kernel (DPK) scaling accuracy was investigated in various homogeneous tissue media. Methods: Using GEometry ANd Tracking 4 Monte Carlo code, DPKs were generated for 5, 8 MeV monoenergetic α particles and 223Ra, 225Ac, and 227Th. Dose was scored in 1 µm thick concentric shells and DPKs were scaled based on the tissue's mass density and compared with the water DPK. Results: Scaled kernels agreed within ±5% except near the Bragg peaks, where they differed up to 25%. Conclusions: The authors conclude that kernel scaling based on mass density of the transport medium can be utilized accurately up to 5%, excluding Bragg peak regions.

Characterizing a PTW microDiamond detector in kilovoltage radiation beams.

PURPOSE: The aim of this work was to characterize the dosimetric properties of the PTW microDiamond (60019) single crystal synthetic diamond detector (DD) in kilovoltage x-ray beams. The following characteristics were addressed in this study: required preirradiation dose, dose-rate linearity, energy dependence, and percent depth dose response of the DD. METHODS: UWADCL x-ray beams, characterized by NIST-traceable ionization chambers, were used in this study. Preirradiation dose required by the DD, in order to stabilize the detector's response to within 0.1%, was quantitated. Dose-rate dependence was also investigated using the UW250-M and UW50-M beams, where the dose rate was varied by changing the tube current. N k and N D , w calibration coefficients for all the available M series beams at UWADCL were obtained to determine the energy dependence of the DD, Diode E, Diode P, and P11 parallel-plate ionization chamber. A custom-built water tank was utilized to measure the percent depth dose (PDD) response of the DD, Diode E, Diode P, and P11 chamber in UW250-M, UW100-M, and UW50-M beams. The measured PDD response of the detectors was compared with the simulated PDD data using EGSnrc Monte Carlo code. RESULTS: A 1.5 Gy dose-to-water or air-kerma was found to be sufficient for the given DD's response to stabilize to within 0.1% in all of the beams used in this study. The dose-rate dependence parameter, Δ, was found to be 1.00 ± 0.02 and 1.016 ± 0.05 for the UW250-M and UW50-M beams, respectively. Relative to the 60 Co calibration coefficients, the DD was found to under-respond relative to calculated absorbed dose to water response and over-respond relative to the calculated air-kerma response in the M-series beams. Agreement of 1.5% was found between the measured PDD values and Monte Carlo simulated PDD values for UW250-M, UW100-M, and UW50-M beams. CONCLUSIONS: In order to stabilize the response, the DD needs a preirradiation dose, which is unique to every DD. A linear relationship between detector response and dose rate was found within the evaluated uncertainty. An energy dependence of the DD was studied, which is more pronounced in the low-energy beams and can be partially attributed to the metal contact material around the sensitive volume of the DD. Overall, the DD was found to be suitable for kilovoltage x-ray dosimetry.