An EGS Monte Carlo model for Varian TrueBEAM treatment units: Commissioning and experimental validation of source parameters

Next generation high resolution perovskite direct conversion detector: Monte Carlo design optimisation and virtual clinical trial

We implement virtual clinical integration of next-generation perovskite detectors into common x-ray imaging devices. This was achieved by performing Monte Carlo (MC) optimisation of the design and benchmarking of low cost, high spatial resolution, direct conversion perovskite crystal x-ray flat panel imagers for a next generation of breast-, MV-, and kV-cone beam CT detectors. Semiconductor methylammonium lead bromide perovskite crystals energy deposition efficiencies calculated in TOPAS were benchmarked against four common detector materials for twelve detector crystal thicknesses between 40 to 15 mm and ten beam energies ranging from 20 keV to 6 MeV. Based on these simulations, Koning's dedicated breast CT, and Varian's Truebeam kV- and MV-cone beam CT systems were designated as suitable applications for perovskite detectors. System specific Fastcat hybrid MC cone beam CT image simulation was subsequently used to optimise the perovskite detector design and conduct virtual clinical trials. Device-specific optimal perovskite crystal thicknesses were calculated to be 0.30, 0.86, and 1.99 mm for Koning breast CT and Truebeam kV- and MV-cone beam CT systems, respectively. Replacing the current detectors on these machines with low cost perovskite crystal detectors could be advantageous as it would simultaneously yield 12.1%, 9.5% and 86.1% improvements in detective quantum efficiency as well as increases in contrast to noise ratio in brain, lung, and bone tissues.

A multivariate approach to determine electron beam parameters for a Monte Carlo 6 MV Linac model: Statistical and machine learning methods

PURPOSE

This study aimed to determine the optimal initial electron beam parameters of a Linac for radiotherapy with a multivariate approach using statistical and machine-learning tools.

METHODS

For MC beam commissioning, a 6 MV Varian Clinac was simulated using the Geant4 toolkit. The authors investigated the relations between simulated dose distribution and initial electron beam parameters, namely, mean energy (E), energy spread (ES), and radial beam size (RS). The goodness of simulation was evaluated by the slope of differences between the simulated and the golden beam data. The best-fit combination of the electron beam parameters that minimized the slope of dose difference was searched through multivariate methods using conventional statistical methods and machine-learning tools of the scikit-learn library.

RESULTS

Simulation results with 87 combinations of the electron beam parameters were analyzed. Regardless of being univariate or multivariate, traditional statistical models did not recommend a single parameter set simultaneously minimizing slope of dose differences for percent depth dose (PDD) and lateral dose profile (LDP). Two machine learning classification modules, RandomForestClassifier and BaggingClassifier, agreed in recommending (E = 6.3 MeV, ES = ±5.0%, RS = 1.0 mm) for predicting simultaneous acceptance of PDD and LDP.

CONCLUSIONS

The machine learning with random-forest and bagging classifier modules recommended a consistent result. It was possible to draw an optimal electron beam parameter set using multivariate methods for MC simulation of a radiotherapy 6 MV Linac.

Monte Carlo verification of output correction factors for a TrueBeam STx®

The recent publication of the new code of practice IAEA/AAPM TRS-483 introduces output correction factors to correct detector response changes in relative dosimetry of small photon beams. In TRS-483, average correction factors are reported for several detectors in high-energy photon beams at 6 and 10 MV with and without flattening filter. These correction factors were determined by Monte Carlo simulation or experimental measurements using several linacs of different brands and vendors. The goal of this work was to validate the output correction factors reported in TRS-483 for 6 MV photon beams of a TrueBeam STx® linac. The validation was performed using Monte Carlo simulations of four radiation detectors employed in the dosimetry of small photon beams and whose output correction factors were determined using a different radiation source than TrueBeam STx®. The results show that Monte Carlo calculated output correction factors, and those reported in the code of practice TRS-483 fully agree within ∼1%. The use of generic correction factors for a TrueBeam STx® and the detectors studied in this work is suitable for small field dosimetry static beams within the uncertainties of Monte Carlo calculations and output correction factors reported in TRS-483.

Monte Carlo commissioning of radiotherapy LINAC-Introducing an improved methodology

Purpose

Monte Carlo (MC) commissioning of medical linear accelerator (LINAC) is a time-consuming process involving a comparison between measured and simulated cross beam/lateral profiles and percentage depth doses (PDDs) for various field sizes. An agreement between these two data sets is sought by trial and error method while varying the incident electron beam parameters, such as electron beam energy or width, etc. This study aims to improve the efficiency of MC commissioning of a LINAC by assessing the feasibility of using a limited number of simulated PDDs.

Materials and methods

Using EGSnrc codes, a Varian Clinac 2100 unit has been commissioned for 6 MV photon beam, and a methodology has been proposed to identify the incident electron beam parameters in a speedier fashion. Impact of voxel size in 3-dimensions and cost functions used for comparison of the measured and simulated data have been investigated along with the role of interpolation.

Results

A voxel size of 1 × 1×0.5 cm³ has been identified as suitable for accurate and fast commissioning of the LIANC. The optimum number of simulated PDDs (required for further interpolation) has been found to be five.

Conclusion

The present study suggests that PDDs alone at times can be insufficient for an unambiguous commissioning process and should be supported by including the lateral beam profiles in the process.