A fast jaw-tracking model for VMAT and IMRT Monte Carlo simulations.

Modern radiotherapy techniques involve routine use of volumetric arc therapy (VMAT) and intensity modulated radiotherapy (IMRT) with jaw-tracking - dynamic motion of the secondary collimators (jaws) in tandem with multi-leaf collimators (MLCs). These modalities require accurate dose calculations for the purposes of treatment planning and dose verification. Monte Carlo (MC) methods for radiotherapy dose calculation are widely accepted as capable of achieving high accuracy. This paper presents an efficiency-enhancement method for secondary collimator modeling, presented in the context of a tool for MC-based dose second checks. The model constitutes an accuracy trade-off in the source model for the sake of efficiency enhancement, but maintains the advantages of MC transport in patient heterogeneities. The secondary collimator model is called Flat-Absorbing-Jaw-Tracking (FAJT). Transmission through and scatter from the secondary collimators is neglected, and jaws are modeled as perfectly absorbing planes. To couple the motion of secondary collimators with MLCs for jaw-tracking, the FAJT model was built into the VCU-MLC model. Gamma-index analysis of the dose distributions from FAJT against the full BEAMnrc MC simulations showed over 99% pass rate for a range of open fields, two clinical IMRT, and one VMAT treatment plan, for 2%/2 mm criteria above 10%. Using FAJT, the simulation speed of the secondary collimators for open fields increased by a factor of 237, 1489, and 1395 for 4 × 4, 10 × 10, and 30 × 30 cm2 , respectively. In general, clinically oriented simulation times are reduced from "hours" to "minutes" on identical hardware. Results for nine representative clinical cases (seven with jaw-tracking) are presented. The average 2%/2 mm γ-test success rate above the 80% isodose was 96.8% when tested against the EPIDose electronic portal image-based dose reconstruction method and 97.3% against the Eclipse analytical anisotropic algorithm.

Comparison of measured Varian Clinac 21EX and TrueBeam accelerator electron field characteristics.

Dosimetric comparisons of radiation fields produced by Varian's newest linear accelerator, the TrueBeam, with those produced by older Varian accelerators are of interest from both practical and research standpoints. While photon fields have been compared in the literature, similar comparisons of electron fields have not yet been reported. In this work, electron fields produced by the TrueBeam are compared with those produced by Varian's Clinac 21EX accelerator. Diode measurements were taken of fields shaped with electron applicators and delivered at 100 cm SSD, as well as those shaped with photon MLCs without applicators and delivered at 70 cm SSD for field sizes ranging from 5 × 5 to 25 × 25 cm² at energies between 6 and 20 MeV. Additionally, EBT2 and EBT3 radio-chromic film measurements were taken of an MLC-shaped aperture with closed leaf pairs delivered at 70 cm SSD using 6 and 20 MeV electrons. The 6 MeV fields produced by the TrueBeam and Clinac 21EX were found to be almost indistinguishable. At higher energies, TrueBeam fields shaped by electron applicators were generally flatter and had less photon contamination compared to the Clinac 21EX. Differences in PDDs and profiles fell within 3% and 3 mm for the majority of measurements. The most notable differences for open fields occurred in the profile shoulders for the largest applicator field sizes. In these cases, the TrueBeam and Clinac 21EX data differed by as much as 8%. Our data indicate that an accurate electron beam model of the Clinac 21EX could be used as a starting point to simulate electron fields that are dosimetrically equivalent to those produced by the TrueBeam. Given that the Clinac 21EX shares head geometry with Varian's iX, Trilogy, and Novalis TX accelerators, our findings should also be applicable to these machines.

Comparative evaluation of modern dosimetry techniques near low- and high-density heterogeneities.

The purpose of this study is to compare performance of several dosimetric meth-ods in heterogeneous phantoms irradiated by 6 and 18 MV beams. Monte Carlo (MC) calculations were used, along with two versions of Acuros XB, anisotropic analytical algorithm (AAA), EBT2 film, and MOSkin dosimeters. Percent depth doses (PDD) were calculated and measured in three heterogeneous phantoms. The first two phantoms were a 30 × 30 × 30 cm3 solid-water slab that had an air-gap of 20× 2.5 × 2.35 cm3. The third phantom consisted of 30 × 30 × 5 cm3 solid water slabs, two 30 × 30 × 5 cm3 slabs of lung, and one 30 × 30 × 1 cm3 solid water slab. Acuros XB, AAA, and MC calculations were within 1% in the regions with particle equilibrium. At media interfaces and buildup regions, differences between Acuros XB and MC were in the range of +4.4% to -12.8%. MOSkin and EBT2 measurements agreed to MC calculations within ~ 2.5%, except for the first cen-timeter of buildup where differences of 4.5% were observed. AAA did not predict the backscatter dose from the high-density heterogeneity. For the third, multilayer lung phantom, 6 MV beam PDDs calculated by all TPS algorithms were within 2% of MC. 18 MV PDDs calculated by two versions of Acuros XB and AAA differed from MC by up to 2.8%, 3.2%, and 6.8%, respectively. MOSkin and EBT2 each differed from MC by up to 2.9% and 2.5% for the 6 MV, and by -3.1% and ~2% for the 18 MV beams. All dosimetric techniques, except AAA, agreed within 3% in the regions with particle equilibrium. Differences between the dosimetric techniques were larger for the 18 MV than the 6 MV beam. MOSkin and EBT2 measurements were in a better agreement with MC than Acuros XB calculations at the interfaces, and they were in a better agreement to each other than to MC. The latter is due to their thinner detection layers compared to MC voxel sizes.

Characterization of a 0.8 mm3Medscint plastic scintillator detector system for small field dosimetry.

Objective. Plastic scintillator detectors (PSDs) have demonstrated ability to meet requirements of small field dosimetry. Medscint developed a 1 mm long, 1 mm diameter cylindrical PSD with effective volume of 0.8 mm3. Clinically relevant, small field dosimetric properties of this detector, combined with a novel scintillation dosimetry system-HYPERSCINT RP-200, and HYPERDOSE analysis software were evaluated in this study.Approach. This novel scintillator-based dosimetry system was characterized with 6 MV-WFF and 10 MV-FFF x-ray beams delivered by Varian TrueBeamTMlinear accelerator. The detector was characterized for leakage, short-term repeatability, dose response linearity, angular response, dose rate response, and field size dependence for radiation field sizes of 0.25 × 0.25 to 10 × 10 cm2. Measured detector specific output ratios were compared with microDiamond output factors to determine small field output correction factors,kQclin,Qmsrfclin,fmsr.Main results. The dosimetry system showed excellent short-term repeatability with standard deviation of only 0.04 ± 0.01%. It demonstrated good dose linearity with variations less than 1.0% for 14.4 cGy and above. The dosimetry system was found to be independent of dose rate and angle of irradiation, with deviations for both below 0.5%. Leakage was found to be comparable to background readings. For 6 MV-WFF energy beams, detector specific output ratios for field sizes down to 1 × 1 cm2agreed with output factors measured with PTW TN60019 microDiamond, thus,kQclin,Qmsrfclin,fmsrequates to unity for these field sizes. For 10 MV-FFF energy beams, detector specific output ratios for field sizes down to 2 × 2 cm2agreed with PTW TN60019 microDiamond output factors, thus,kQclin,Qmsrfclin,fmsrequates to unity for these field sizes.kQclin,Qmsrfclin,fmsrfor field sizes down to 0.5 × 0.5 cm2were determined to be within 6% of unity for both 6 MV-WFF and 10 MV-FFF energy beams.Significance. The HYPERSCINT RP-200 dosimetry system coupled with a 0.8 mm3PSD showed excellent dosimetric properties and was found to be clinically relevant for relative dosimetry down to field sizes of 0.5 × 0.5 cm2and potentially smaller.

Coplanar versus noncoplanar intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT) treatment planning for fronto-temporal high-grade glioma.

The purpose of this study was to compare dosimetric and radiobiological parameters of treatment plans using coplanar and noncoplanar beam arrangements in patients with fronto-temporal high-grade glioma (HGG) generated for intensity-modulated radiotherapy (IMRT) or volumetric-modulated arc therapy (VMAT). Ten cases of HGG overlapping the optic apparatus were selected. Four separate plans were created for each case: coplanar IMRT, noncoplanar IMRT (ncIMRT), VMAT, and noncoplanar VMAT (ncVMAT). The prescription dose was 60 Gy in 30 fractions. Dose-volume histograms and equivalent uniform doses (EUD) for planning target volumes (PTVs) and organs at risk (OARs) were generated. The four techniques resulted in comparable mean, minimum, maximum PTV doses, and PTV EUDs (p ≥ 0.33). The mean PTV dose and EUD averaged for all techniques were 59.98 Gy (Standard Deviation (SD) ± 0.15) and 59.86 Gy (SD ± 0.27). Non-coplanar IMRT significantly reduced contralateral anterior globe EUDs (6.7 Gy versus 8.2 Gy, p = 0.05), while both ncIMRT and ncVMAT reduced contralateral retina EUDs (16 Gy versus 18.8 Gy, p = 0.03). Noncoplanar techniques resulted in lower contralateral temporal lobe dose (22.2 Gy versus 24.7 Gy). Compared to IMRT, VMAT techniques required fewer monitor units (755 vs. 478, p ≤ 0.001) but longer optimization times. Treatment delivery times were 6.1 and 10.5 minutes for coplanar and ncIMRT versus 2.9 and 5.0 minutes for coplanar and ncVMAT. In this study, all techniques achieved comparable target coverage. Superior sparing of contralateral optic structures was seen with ncIMRT. The VMAT techniques reduced treatment delivery duration but prolonged plan optimization times, compared to IMRT techniques. Technique selection should be individualized, based on patient-specific clinical and dosimetric parameters.

Understanding the impact of RapidArc therapy delivery errors for prostate cancer.

The purpose of this study is to simulate random and systematic RapidArc delivery errors for external beam prostate radiotherapy plans in order to determine the dose sensitivity for each error type. Ten prostate plans were created with a single 360° arc. The DICOM files for these treatment plans were then imported into an in-house computer program that introduced delivery errors. Random and systematic gantry position (0.25°, 0.5°, 1°), monitor unit (MU) (1.25%, 2.5%, 5%), and multileaf collimator (MLC) position (0.5, 1, 2 mm) errors were introduced. The MLC errors were either random or one of three types of systematic errors, where the MLC banks moved in the same (MLC gaps remain unchanged) or opposing directions (increasing or decreasing the MLC gaps). The generalized equivalent uniform dose (gEUD) was calculated for the original plan and all treatment plans with errors introduced. The dose sensitivity for the cohort was calculated using linear regression for the gantry position, MU, and MLC position errors. Because there was a large amount of variability for systematic MLC position errors, the dose sensitivity of each plan was calculated and correlated with plan MU, mean MLC gap, and the percentage of MLC leaf gaps less than 1 and 2 cm for each individual plan. We found that random and systematic gantry position errors were relatively insignificant (< 0.1% gEUD change) for gantry errors up to 1°. Random MU errors were also insignificant, and systematic MU increases caused a systematic increase in gEUD. For MLC position errors, random MLC errors were relatively insignificant up to 2 mm as had been determined in previous IMRT studies. Systematic MLC shift errors caused a decrease of approximately -1% in the gEUD per mm. For systematic MLC gap open errors, the dose sensitivity was 8.2%/mm and for MLC gap close errors the dose sensitivity was -7.2%/mm. There was a large variability for MLC gap open/close errors for the ten RapidArc plans which correlated strongly with MU, mean gap width, and percentage of MLC gaps less than 1 or 2cm. This study evaluates the magnitude of various simulated RapidArc delivery errors by calculating gEUED on various prostate plans.

IEC accelerator beam coordinate transformations for clinical Monte Carlo simulation from a phase space or full BEAMnrc particle source.

Monte Carlo simulation of clinical treatment plans require, in general, a coordinate transformation to describe the incident radiation field orientation on a patient phantom coordinate system. The International Electrotechnical Commission (IEC) has defined an accelerator coordinate system along with positive directions for gantry, couch and collimator rotations. In order to describe the incident beam's orientation with respect to the patient's coordinate system, DOSXYZnrc simulations often require transformation of the accelerator's gantry, couch and collimator angles to describe the incident beam. Similarly, versions of the voxelized Monte Carlo code (VMC(++)) require non-trivial transformation of the accelerator's gantry, couch and collimator angles to standard Euler angles α, ß, γ, to describe an incident phase space source orientation with respect to the patient's coordinate system. The transformations, required by each of these Monte Carlo codes to transport phase spaces through a phantom, have been derived with a rotation operator approach. The transformations have been tested and verified against the Eclipse treatment planning system.

Inference of the optimal pretarget electron beam parameters in a Monte Carlo virtual linac model through simulated annealing.

The purpose of this study was to develop an efficient method to determine the optimal intensity distribution of the pretarget electron beam in a Monte Carlo (MC) accelerator model able to most accurately reproduce a set of measured photon field profiles for a given accelerator geometry and nominal photon beam energy. The method has the ability to reduce the number of simulations required to commission a MC accelerator model and has achieved better agreement with measurement than other methods described in literature. The method begins from a cylindrically symmetric pretarget electron beam (radius of 0.5 cm) of uniform intensity. This beam is subdivided into annular regions of fluence for which each region is individually transported through the accelerator head and into a water phantom. A simulated annealing search is then performed to determine the optimal combination of weights of the annular fluences that provide a best match between the measured dose distributions and the weighted sum of annular dose distributions for particular pretarget electron energy. When restricted to Gaussian intensity distributions, the optimization determined an optimal FWHM=1.34 mm for 18.0 MeV electrons, with a RMSE=0.49% on 40 x 40 cm2 lateral profiles. When allowed to deviate from Gaussian intensities a further reduction in RMSE was achieved. For our Clinac 21 EX accelerator MC model (based on the 1996 Varian Oncology Systems, Monte Carlo Project package), the optimal unrestricted intensity distribution was found to be a Gaussian-like solution (18.0 MeV, FWHM= 1.10 mm, 40 x 40 cm2 profile, and RMSE=0.15%) with the presence of an extra focal halo contribution on the order of 10% of the maximum Gaussian intensity. Using the optimally derived intensity, 10 x 10 and 4 x 4 cm2 profiles were found to be in agreement with measurement with a maximum RMSE=0.49%. The optimized Gaussian and unrestricted values of the electron beam FWHM were both within the range of those inferred by focal spot image measurements performed by Jaffray et al. ["X-ray sources of medical linear accelerators: Focal and extra-focal radiation," Med. Phys. 20, 1417-1427 (1993)]. The inference of an extra focal pretarget electron component may be an indicator of a deficiency in the MC model and needs further investigation.

Monte Carlo calculated kilovoltage x-ray arc therapy plans for three lung cancer patients.

Purpose: The intent of this work was to evaluate the ability of our 200 kV kilovoltage arc therapy (KVAT) system to treat realistic lung tumors without exceeding dose constraints to organs-at-risk (OAR).Methods and Materials: Monte Carlo (MC) methods and the McO optimization framework generated and inversely optimized KVAT treatment plans for 3 SABR lung cancer patients. The KVAT system was designed to treat deep-seated lesions with kilovoltage photons. KVAT delivers dose to roughly spherical PTVs and therefore non-spherical PTVs were divided into spherical sub-volumes. A prescription dose of 12 Gy/fx × 4 fractions was planned to 90% of the PTV volume. KVAT plans were compared to VMC++ calculated, 6 MV stereotactic ablative radiotherapy (SABR) treatment plans. Dose distributions, dose volume histograms, gradient index (GI), planned mean doses and plan treatment times were calculated. Dose constraints for organs-at-risk (OAR) were taken from RTOG 101.Results: All plans, with the exception of the rib dose calculated in one of the KVAT plans for a peripheral lesion, were within dose-constraints. In general, KVAT plans had higher planned doses to OARs. KVAT GI values were 5.7, 7.2 and 8.9 and SABR values were 4.6, 4.1, and 4.7 for patient 1, 2 and 3, respectively. KVAT plan treatment times were 49, 65 and 17 min for patients 1, 2 and 3, respectively.Conclusions: Inverse optimization and MC methods demonstrated the ability of KVAT to produce treatment plans without exceeding TG 101 dose constraints to OARs for 2 out of 3 investigated lung cancer patients.

Application of the equivalent uniform stochastic dose (EUSD) to TCP calculations incorporating dose uncertainty and fractionation effects.

A method of tumour control probability (TCP) evaluation that intrinsically accounts for the dose uncertainties is demonstrated in this paper. The dose uncertainty is taken into account through the concept of an equivalent stochastic dose (ESD) defined as the dose to a voxel that results in the mean expected survival fraction from a process randomly depositing a dose D. Applying ESD to a non-uniform dose distribution yields the concept of equivalent uniform stochastic dose (EUSD). TCP is modeled to include dose uncertainty and dose inhomogeneity as TCP (EUSD). It is shown that Webb-Nahum and Niemierko-Goitein TCP models both converge to TCP (EUSD) when dose uncertainty is accounted for. TCPs were calculated for a tumour irradiated with a dose of 60 and 70 Gy at 2 Gy per fraction, where the dose uncertainty in tumour sub-volumes was variable. Effect of the dose fractionation on TCP in the presence of the dose uncertainty was also investigated using TCP (EUSD) model. Tolerance uncertainty on the dose resulting in tolerance TCP loss (assumed as 5%) was calculated for a range of radiobiological parameters. TCP degradation due to the treatment dose uncertainty was evaluated. It is shown that degradation of the TCP is controlled by the voxels where the dose has high uncertainty. For a modeled tumour (alpha=0.3, alpha/beta=10, N0=10(8)) irradiated with 60 Gy, 12% dose uncertainty at 1% fractional tumour sub-volume reduced the TCP from 95% to 50%. Presented TCP (EUSD) model also demonstrated capability to robustly evaluate the loss of TCP due to the dose uncertainties at different fractionation regimens. It is shown that the tolerance uncertainty reduces with decreased number of fractions indicating that hypo-fractionated treatments may require more accurate dose delivery.

Output and ([Formula: see text]) correction factors measured and calculated in very small circular fields for microDiamond and EFD-3G detectors.

The purpose of this work was to obtain [Formula: see text] factors for microDiamond and EFD-3G detectors in very small (less than 5 mm) circular fields. We also investigated the impact of possible variations in microDiamond detector design schematics on the calculated [Formula: see text] factors. Output factors (OF's) of 6 MV beams from TrueBeam linac collimated with 1.27-40 mm diameter cones were measured with EBT3 films, microDiamond and EFD-3G detectors as well as calculated (in water) using Monte Carlo (MC) methods. Based on EBT3 measurements and MC calculations [Formula: see text] factors were derived for these detectors. MC calculations were performed for microDiamond detector in parallel and perpendicular orientations relative to the beam axis. Furthermore, [Formula: see text] factors were calculated for two microDiamond detector models, differing by the presence or absence of metallic pins. The measured OFs agreed within 2.4% for fields ⩾10 mm. For the cones of 1.27, 2.46, and 3.77 mm maximum differences were 17.9%, 1.8% and 9.0%, respectively. MC calculated output factors in water agreed with those obtained using EBT3 film within 2.2% for all fields. MC calculated [Formula: see text] factors for microDiamond detector in fields ⩾10 mm ranged within 0.975-1.020 for perpendicular and parallel orientations. MicroDiamond detector [Formula: see text] factors calculated for the 1.27, 2.46 and 3.77 mm fields were 1.974, 1.139 and 0.982 with detector in parallel orientation, and these factors were 1.150, 0.925 and 0.914 in perpendicular orientation. Including metallic pins in the microDiamond model had little effect on calculated [Formula: see text] factors. EBT3 and MC obtained [Formula: see text] factors agreed within 3.7% for fields of ⩾3.77 mm and within 5.9% for smaller cones. Including metallic pins in the detector model had no effect on calculated [Formula: see text] factors. Our results show that microDiamond and EFD-3G detectors can be used in very small (1.27-3.77 mm) fields once [Formula: see text] corrections determined in this work are applied. Expected uncertainty of such measurements will be in the range of 8%-2.5%.

Inverse optimization of low-cost kilovoltage x-ray arc therapy plans.

PURPOSE: The objective of this work was to investigate the benefits of using inverse optimization treatment planning for kilovoltage arc therapy (KVAT) and to assess the dosimetric limitations of KVAT. METHODS: Monte Carlo (MC) calculated, inversely optimized KVAT plans of spherical, idealized breast, lung, and prostate lesions were calculated using the EGSnrc/BEAMnrc and DOSXYZnrc MC codes. The dose delivered with the KVAT system, which generates 200-225 kV photon beamlets, was calculated and inversely optimized using an optimization framework developed at McGill University. KVAT dose distributions were compared with inversely optimized and MC generated megavoltage (MV) volumetric modulated arc therapy (VMAT) plans as a reference. Prescription doses delivered to 95% of the planning target volume (PTV) were 38.5 (10 fractions), 60 (30 fractions) and 73.8 (41 fractions) Gy for the breast, lung and prostate patients, respectively. Dose distributions, dose volume histograms, and PTV homogeneity indices were used to evaluate KVAT and VMAT plans based on RTOG protocols. RESULTS: All organ-at-risk (OAR) doses were within prescribed dose limits for KVAT and VMAT plans. Generally, KVAT plans delivered higher doses to OARs. For example, due to the lower energy of KVAT, 50% of the rib volume received 12.9 Gy from KVAT while only receiving 2.5 Gy from VMAT. OAR doses were especially high for the KVAT prostate plan due to the presence of large volumes of bony anatomy, which illustrates a limitation of the KVAT system. The KVAT treatment times per fraction for the breast, lung and prostate patients were 2.8, 2.6 and 5.5 min, respectively. CONCLUSIONS: The inversely optimized KVAT plans presented in this work have demonstrated the ability of our novel low-cost, kilovoltage x-ray therapy system to safely treat deep-seated spherical lesions in breast and lung patients while meeting RTOG dose constraints on OARs.

Evaluation of latent variances in Monte Carlo dose calculations with Varian TrueBeam photon phase-spaces used as a particle source.

The purpose of this study was to evaluate the latent variance (LV) of Varian TrueBeam photon phase-space files (PSF) for open 10 × 10 cm2 and small stereotactic fields and estimate the number of phase spaces required to be summed up in order to maintain sub-percent LV in Monte Carlo (MC) dose calculations. BEAMnrc/DOSXYZnrc software was used to transport particles from Varian phase-space files (PSFA) through the secondary collimators. Transported particles were scored into another phase-space located under the jaws (PSFB), or transported further through the cone collimators and scored straight below, forming PSFC. Phase-space files (PSFB) were scored for 6 MV-FFF, 6 MV, 10 MV-FFF, 10 MV and 15 MV beams with 10 × 10 cm2 field size, and PSFC were scored for 6 MV beam under circular cones of 0.13, 0.25, 0.35, and 1 cm diameter. Both PSFB and PSFC were transported into a water phantom with particle recycling number ranging from 10 to 1000. For 10 × 10 cm2 fields 0.5 × 0.5 × 0.5 cm3 voxels were used to score the dose, whereas the dose was scored in 0.1 × 0.1 × 0.5 cm3 voxels for beams collimated with small cones. In addition, for small 0.25 cm diameter cone-collimated 6 MV beam, phantom voxel size varied as 0.02 × 0.02 × 0.5 cm3, 0.05 × 0.05 × 0.5 cm3 and 0.1 × 0.1 × 0.5 cm3. Dose variances were scored in all cases and LV evaluated as per Sempau et al. For the 10 × 10 cm2 fields calculated LVs were greatest at the phantom surface and decreased with depth until they reached a plateau at 5 cm depth. LVs were found to be 0.54%, 0.96%, 0.35%, 0.69% and 0.57% for the 6 MV-FFF, 6 MV, 10 MV-FFF, 10 MV and 15 MV energies, respectively at the depth of 10 cm. For the 6 MV phase-space collimated with cones of 0.13, 0.25, 0.35, 1.0 cm diameter, the LVs calculated at 1.5 cm depth were 75.6%, 25.4%, 17.6% and 8.0% respectively. Calculated LV for the 0.25 cm cone-collimated 6 MV beam were 61.2%, 40.7%, 22.5% in 0.02 × 0.02 × 0.5 cm3, 0.05 × 0.05 × 0.5 cm3 and 0.1 × 0.1 × 0.5 cm3 voxels respectively. In order to achieve sub-percent LV in open 10 × 10 cm2 field MC simulations a single PSF can be used, whereas for small SRS fields (0.13-1.0 cm) more PSFs (66-8 PSFs) would have to be summed.

Monte Carlo simulations of a kilovoltage external beam radiotherapy system on phantoms and breast patients.

PURPOSE: To determine the most suitable lesion size and depth for radiotherapy treatments with a prototype kilovoltage x-ray arc therapy (KVAT) system through Monte Carlo simulations of the dose delivered to lesion, dose homogeneity, and lesion-to-skin ratio. METHODS: Monte Carlo simulations were used to calculate dose distributions generated by a novel low-energy kilovoltage x-ray system to a variety of clinically relevant lesion sizes and depths in phantoms and for hypothetical partial breast irradiations of patients in supine and prone positions. The treatments by 200 kV KVAT system were modeled for four sizes of tumor (1-4 cm diameter) at three depths (superficial, middle, and deep) in two sizes of cylindrical water phantoms (16.2-cm and 32.2-cm diameter). In addition, treatments of 3-cm and 4-cm diameter lesions were modeled for two breast patients in prone and supine positions. Dose distributions were calculated using the EGSnrc/DOSXYZnrc code package. Phantom study metrics included lesion-to-skin ratio, dose delivered to isocenter (cGy/min), dose homogeneity, dose profiles, and cumulative dose volume histograms. Lesion-to-skin ratio, lesion-to-rib ratio, dose profiles, and cumulative dose volume histograms were used to evaluate simulated breast patient treatments. Supine breast irradiations were compared to 6-MV VMAT plans. The criterion applied to evaluate the dose distributions was derived from NSABP-B39/RTOG 0413 for accelerated partial breast irradiation. Skin dose was limited to a maximum of 250 cGy for a prescribed lesion dose of 385 cGy per fraction (with the whole treatment being delivered in 10 fractions). This produced the minimum lesion-to-skin dose ratio of 1.5 that served as the main guideline, along with other metrics, for evaluation of future clinical viability of treatments. RESULTS: Phantom dose distributions in the centrally located lesions treated with 360-degree KVAT were found to be superior to dose distributions in off-center lesions with the exception of isocenter dose, which was highest for lesions located closer to the phantom surface. Dose metrics were more favorable for smaller lesions, suggesting that KVAT might be most suitable for treatment of lesions of 1-2 cm in diameter down to depths of 8.1 cm along with 3 cm lesions at depths from 3 cm to 8.1 cm. In addition, treatments of 4-cm lesions were found to be acceptable down to the depths of 4.1 cm (in the 16.2-cm phantom) and 8.1 cm (in the 32.2-cm phantom). At depths from 8.1-cm to 16.1-cm, treatments of 1-cm to 4-cm lesions are possible at the cost of decreased dose rate. KVAT breast treatments in the supine patient position demonstrated that increasing the arc angle and decreasing lesion size improved lesion-to-skin ratio and lesion-to-rib ratio. Supine breast data indicate that 3-cm lesions are treatable at a minimum depth of 3 cm. The 6-MV VMAT plan resulted in lower doses to the ipsilateral lung and the body, but a higher heart dose compared to the KVAT plans. Dose distributions for the prone breast phantoms were superior to the supine cases due to the increased treatment angle of 360-degrees. CONCLUSIONS: Although nonoptimized KVAT dose distributions presented here were of inferior quality to VMAT plans, this work has demonstrated the feasibility of delivering low-energy kilovoltage x-rays to lesions up to 4 cm in diameter to depths of 8.1 cm while sparing surrounding tissue.

Evaluation of the analytical anisotropic algorithm in an extreme water-lung interface phantom using Monte Carlo dose calculations.

Our study compares the performance of the analytical anisotropic algorithm (AAA), a new superposition-convolution algorithm recently implemented in the Eclipse (Varian Medical Systems, Palo Alto, CA) Integrated Treatment Planning System (TPS), to that of the pencil beam convolution (PBC) algorithm in an extreme (C-shaped, horizontal and vertical boundaries) water-lung interface phantom. Monte Carlo (MC) calculated dose distributions for a variety of clinical beam configurations at nominal energies of 6-MV and 18-MV are used as benchmarks in the comparison. Dose profiles extracted at three depths (4, 10, and 16 cm), two-dimensional (2D) maps of the dose differences, and dose difference statistics are used to quantify the accuracy of both photon-dose calculation algorithms. Results show that the AAA is considerably more accurate than the PBC, with the standard deviation of the dose differences within a region encompassing the lung block reduced by a factor of 2 and more. Confidence limits with the AAA were 4% or less for all beam configurations investigated; with the PBC, confidence limits ranged from 3.5% to 11.2%. Finally, AAA calculations for the small 4 x 4 18-MV beam, which is poorly modeled by PBC (dose differences as high as 16.1%), provided the same accuracy as the PBC model of the 6-MV beams commonly acceptable in clinical situations.

Validation of Varian TrueBeam electron phase-spaces for Monte Carlo simulation of MLC-shaped fields.

PURPOSE: This work evaluates Varian's electron phase-space sources for Monte Carlo simulation of the TrueBeam for modulated electron radiation therapy (MERT) and combined, modulated photon and electron radiation therapy (MPERT) where fields are shaped by the photon multileaf collimator (MLC) and delivered at 70 cm SSD. METHODS: Monte Carlo simulations performed with EGSnrc-based BEAMnrc/DOSXYZnrc and penelope-based PRIMO are compared against diode measurements for 5 × 5, 10 × 10, and 20 × 20 cm(2) MLC-shaped fields delivered with 6, 12, and 20 MeV electrons at 70 cm SSD (jaws set to 40 × 40 cm(2)). Depth dose curves and profiles are examined. In addition, EGSnrc-based simulations of relative output as a function of MLC-field size and jaw-position are compared against ion chamber measurements for MLC-shaped fields between 3 × 3 and 25 × 25 cm(2) and jaw positions that range from the MLC-field size to 40 × 40 cm(2). RESULTS: Percent depth dose curves generated by BEAMnrc/DOSXYZnrc and PRIMO agree with measurement within 2%, 2 mm except for PRIMO's 12 MeV, 20 × 20 cm(2) field where 90% of dose points agree within 2%, 2 mm. Without the distance to agreement, differences between measurement and simulation are as large as 7.3%. Characterization of simulated dose parameters such as FWHM, penumbra width and depths of 90%, 80%, 50%, and 20% dose agree within 2 mm of measurement for all fields except for the FWHM of the 6 MeV, 20 × 20 cm(2) field which falls within 2 mm distance to agreement. Differences between simulation and measurement exist in the profile shoulders and penumbra tails, in particular for 10 × 10 and 20 × 20 cm(2) fields of 20 MeV electrons, where both sets of simulated data fall short of measurement by as much as 3.5%. BEAMnrc/DOSXYZnrc simulated outputs agree with measurement within 2.3% except for 6 MeV MLC-shaped fields. Discrepancies here are as great as 5.5%. CONCLUSIONS: TrueBeam electron phase-spaces available from Varian have been implemented in two distinct Monte Carlo simulation packages to produce dose distributions and outputs that largely reflect measurement. Differences exist in the profile shoulders and penumbra tails for the 20 MeV phase-space off-axis and in the outputs for the 6 MeV phase-space.

Measured and Monte Carlo simulated electron backscatter to the monitor chamber for the Varian TrueBeam Linac.

To accurately simulate therapeutic electron beams using Monte Carlo methods, backscatter from jaws into the monitor chamber must be accounted for via the backscatter factor, S b. Measured and simulated values of S b for the TrueBeam are investigated. Two approaches for measuring S b are presented. Both require service mode operation with the dose and pulse forming networking servos turned off in order to assess changes in dose rate with field size. The first approach samples an instantaneous dose rate, while the second approach times the delivery of a fixed number of monitor units to assess dose rate. Dose rates were measured for 6, 12 and 20 MeV electrons for jaw- or MLC-shaped apertures between [Formula: see text] and [Formula: see text] cm2. The measurement techniques resulted in values of S b that agreed within 0.21% for square and asymmetric fields collimated by the jaws. Measured values of S b were used to calculate the forward dose component in a virtual monitor chamber using BEAMnrc. Based on this forward component, simulated values of S b were calculated and compared to measurement and Varian's VirtuaLinac simulations. BEAMnrc results for jaw-shaped fields agreed with measurements and with VirtuaLinac simulations within 0.2%. For MLC-shaped fields, the respective measurement techniques differed by as much as 0.41% and BEAMnrc results differed with measurement by as much as 0.4%, however, all measured and simulated values agreed within experimental uncertainty. Measurement sensitivity was not sufficient to capture the small backscatter effect due to the MLC, and Monte Carlo predicted backscatter from the MLC to be no more than 0.3%. Backscatter from the jaws changed the electron dose rate by up to 2.6%. This reinforces the importance of including a backscatter factor in simulations of electron fields shaped with secondary collimating jaws, but presents the option of ignoring it when jaws are retracted and collimation is done with the MLC.

Modelling the variation in rectal dose due to inter-fraction rectal wall deformation in external beam prostate treatments.

Prostate radiotherapy inevitably deposits radiation dose in the rectal wall, and the dose delivered to prostate is limited by the expected rectal complications. Accurate evaluation of the rectal dose is non-trivial due to a number of factors. One of these is variation of the shape and position of the rectal wall (with respect to the clinical target volume (CTV)), which may differ daily from that taken during planning CT acquisition. This study uses data currently available in the literature on rectal wall motion to provide estimates of mean population rectal wall dose. The rectal wall geometry is characterized by a population mean radius of the rectum as well as inter-patient and inter-fraction standard deviations in rectum radius. The model is used to evaluate the range of inter-fraction and inter-patient rectal dose variations. The simulation of individual patients with full and empty rectum in the planning CT scan showed that large variations in rectal dose (>15 Gy) are possible. Mean calculated dose accounting for treatment and planning uncertainties in the rectal wall surface was calculated as well as the map of planning dose over/underpredictions. It was found that accuracy of planning dose is dependent on the CTV-PTV margin size with larger margins producing more accurate estimates. Over a patient population, the variation in rectal dose is reduced by increasing the number of pre-treatment CT scans.

A hybrid phase-space and histogram source model for GPU-based Monte Carlo radiotherapy dose calculation.

In GPU-based Monte Carlo simulations for radiotherapy dose calculation, source modelling from a phase-space source can be an efficiency bottleneck. Previously, this has been addressed using phase-space-let (PSL) sources, which provided significant efficiency enhancement. We propose that additional speed-up can be achieved through the use of a hybrid primary photon point source model combined with a secondary PSL source. A novel phase-space derived and histogram-based implementation of this model has been integrated into gDPM v3.0. Additionally, a simple method for approximately deriving target photon source characteristics from a phase-space that does not contain inheritable particle history variables (LATCH) has been demonstrated to succeed in selecting over 99% of the true target photons with only ~0.3% contamination (for a Varian 21EX 18 MV machine). The hybrid source model was tested using an array of open fields for various Varian 21EX and TrueBeam energies, and all cases achieved greater than 97% chi-test agreement (the mean was 99%) above the 2% isodose with 1% / 1 mm criteria. The root mean square deviations (RMSDs) were less than 1%, with a mean of 0.5%, and the source generation time was 4-5 times faster. A seven-field intensity modulated radiation therapy patient treatment achieved 95% chi-test agreement above the 10% isodose with 1% / 1 mm criteria, 99.8% for 2% / 2 mm, a RMSD of 0.8%, and source generation speed-up factor of 2.5.

Pre-treatment radiotherapy dose verification using Monte Carlo doselet modulation in a spherical phantom.

Due to the increasing complexity of radiotherapy delivery, accurate dose verification has become an essential part of the clinical treatment process. The purpose of this work was to develop a pre-treatment verification technique capable of quickly reconstructing 3D dose distributions from both coplanar and non-coplanar treatments. For each treatment field, electronic portal images were taken in non-transmission mode (with no patient in the beam) allowing the derivation of the delivered fluence maps. The dose reconstruction was then performed in a spherical water phantom by modulating and summing the Monte Carlo (MC) doselets, defined on a spherical co-ordinate system, and pre-calculated from azimuthally symmetric fluence above the jaws. The technique, called the spherical doselet modulation (SDM) method, essentially eliminates the statistical uncertainty of the MC dose calculations by exploiting the azimuthal symmetry in both a patient-independent phase-space and in a virtual spherical water phantom. For example, this symmetry allowed the number of doselets necessary for dose reconstruction to be reduced by a factor of ∼250. In this work, only 51 radially binned doselets were used (each generated from all particles in a given annulus of the phase-space, azimuthally redistributed into a small cylindrical sector). The SDM method mitigates the most computationally intensive part of this type of dose reconstruction--reading, weighting and summing dose matrices. The accuracy of the system was tested against MC calculations as well as our previously reported phase-space modulation method, using a series of open field and IMRT cases. The mean chi- and gamma-test 3%/3 mm success rates of the SDM method were 98.6% and 99.5%, respectively, when compared to full MC simulation. The total calculation time was 96 s per treatment field on a single processor core.