Determination of initial electron parameters by means of Monte Carlo simulations for the Siemens Artiste Linac 6 MV photon beam

Sensitivity Analysis of a 6 MeV Photon Beam Monte Carlo Model

Background

This study aimed to optimize efficiency in Monte Carlo (MC) simulation using sensitivity analysis of a beam model.

Methods

The BEAMnrc-based model of 6 MV beam of a Siemens Primus linac was developed. For sensitivity analysis, the effect of the electron source, treatment head, and virtual phantom specifications on calculated percent depth dose (PDD) and lateral dose profiles was evaluated.

Results

The optimum mean energy (E) and the full width at half maximum (FWHM) of the intensity distribution of the electron beam were calculated as 6.7 MeV and 3 mm, respectively. Increasing E from 6.1 to 6.7 MeV, increased the PDD in the fall-off region by 4.70% and decreased the lateral profile by 8.76%. Changing the FWHM had a significant effect on the buildup region of PDD and the horns and out-of-field regions of the lateral profile. Increasing the collimators opening by 0.5 mm, PDD increased by 2.13% and the central and penumbra regions of profiles decreased by 1.98% and 11.40% respectively. Collimator properties such as thickness and density were effective in changing the penumbra (11.32% for 0.25 cm increment) and the out-of-field (22.82% for 3 g/cm³) regions of the lateral profiles.

Conclusion

Analysis of a 6 MV model showed that PDD profiles were more sensitive to changes in energy than to FWHM of the electron source. The lateral profiles were sensitive to E, FWHM, and collimator opening. The density of the collimator affected only the out-of-field region of lateral profiles. The findings of this study may be used to make benchmarking of an MC beam model more efficient.

The Study of Dosimetric Characteristics of the XHA600D Medical Linear Accelerator Based on a Monte Carlo Code

By investigating the influence of initial electrons on dosimetric characteristics, reasonable incident electron parameters for the nominal 6 MV photon beam of the XHA600D accelerator are finally established, i.e., a 6 MeV monoenergetic electron beam with a radial intensity FWHM of 2.5 mm and an angular divergency of 0.15°. Based on reasonable initial parameters, Percentage Depth Doses (PDDs), Off-Axis Ratios (OARs), total scatter factors, beam qualities, and penumbra widths of both flatteningfilter (FF) and flattening-filter-free (FFF) beams for fields ranging from 4 × 4 to 30 × 30 cm2 are simulated systematically with EGSnrc codes. Not only the simulated dosimetric properties are in excellent agreement with the measurements, but also the dosimetric discrepancies between FF and FFF beams are consistent with the laws of previous studies on other accelerators. Therefore, reasonable incident electron parameters are able to accurately verify the performance of the XHA600D accelerator and can be used for further dosimetry research.

Comparison of Monaco treatment planning system algorithms and Monte Carlo simulation for small fields in anthropomorphic RANDO phantom: The esophagus case

Background

In this study, the dose distributions obtained by the algorithms used in Monaco treatment planning system (TPS) and Monte Carlo (MC) simulation were compared for small fields in the anthropomorphic RANDO phantom, and then, the results were analyzed using the gamma analysis method.

Materials and Methods

In the study, dose distributions obtained from the collapse cone algorithm, MC algorithm, and MC simulation were examined. The EGSnrc was utilized for MC simulation.

Results

In radiation fields smaller than 3 cm × 3 cm, the doses calculated by the CC algorithm are particularly high in the region of lung/soft-tissue interfaces. In the region of soft-tissue/vertebral interfaces, the doses calculated by the CC algorithm and the MC algorithm are compatible with the MC simulation. For each algorithm, the main reason for the non-overlapping dose curves in small fields compared to MC simulation is that the lateral electronic equilibrium loss is not taken into account by the algorithms.

Conclusion

The doses calculated by the algorithms used in TPS may differ, especially in environments where density changes are sharp. Even if the radiation dose from different angles is calculated similarly in the target area by the algorithms, the calculated doses in the tissues in each radiation field path may be different. Therefore, to increase the quality of radiotherapy and to protect critical organs more accurately, the accuracy of the algorithms in TPS should be checked before treatment, especially in multi-field treatments such as stereotactic body radiation therapy and intensity-modulated radiotherapy for tumors in the abdominal region.

The Effect of Algorithms on Dose Distribution in Inhomogeneous Phantom: Monaco Treatment Planning System versus Monte Carlo Simulation

Background:

The aim of this study is to evaluate the dose calculation algorithms commonly used in TPS by using MC simulation in the highly different inhomogeneous region and in the small fields and to provide the following uniquely new information in the study of correction algorithm.

Materials and Methods:

We compared the dose distribution obtained by Monaco TPS for small fields.

Results:

When we examine lung medium, for four different fields, we can see that the algorithms begin to differ. In both the lung and bone environment, the percentage differences decrease as the field size increases. In areas less than or equal to 3x3 cm2, there are serious differences between the algorithms. The CC algorithm calculates a low dose value as the photon passes from the lung environment to water environment. We can also see that this algorithm measures a low dose value in voxel as the photon passes from the water medium to the bone medium. In the transition from the water environment to the bone environment or from the bone environment to the water environment, the results of the CC algorithm are not close to MC simulation.

Conclusion:

The effect of the algorithms used in TPS on dose distribution is very strong, especially in environment with high electron density variation and in applications such as Stereotactic Body Radiotherapy and Intensity Modulated Radiotherapy where small fields are used.

DeepBeam: a machine learning framework for tuning the primary electron beam of the PRIMO Monte Carlo software

Background

Any Monte Carlo simulation of dose delivery using medical accelerator-generated megavolt photon beams begins by simulating electrons of the primary electron beam interacting with a target. Because the electron beam characteristics of any single accelerator are unique and generally unknown, an appropriate model of an electron beam must be assumed before MC simulations can be run. The purpose of the present study is to develop a flexible framework with suitable regression models for estimating parameters of the model of primary electron beam in simulators of medical linear accelerators using real reference dose profiles measured in a water phantom.

Methods

All simulations were run using PRIMO MC simulator. Two regression models for estimating the parameters of the simulated primary electron beam, both based on machine learning, were developed. The first model applies Principal Component Analysis to measured dose profiles in order to extract principal features of the shapes of the these profiles. The PCA-obtained features are then used by Support Vector Regressors to estimate the parameters of the model of the electron beam. The second model, based on deep learning, consists of a set of encoders processing measured dose profiles, followed by a sequence of fully connected layers acting together, which solve the regression problem of estimating values of the electron beam parameters directly from the measured dose profiles. Results of the regression are then used to reconstruct the dose profiles based on the PCA model. Agreement between the measured and reconstructed profiles can be further improved by an optimization procedure resulting in the final estimates of the parameters of the model of the primary electron beam. These final estimates are then used to determine dose profiles in MC simulations.

Results

Analysed were a set of actually measured (real) dose profiles of 6 MV beams from a real Varian 2300 C/D accelerator, a set of simulated training profiles, and a separate set of simulated testing profiles, both generated for a range of parameters of the primary electron beam of the Varian 2300 C/D PRIMO simulator. Application of the two-stage procedure based on regression followed by reconstruction-based minimization of the difference between measured (real) and reconstructed profiles resulted in achieving consistent estimates of electron beam parameters and in a very good agreement between the measured and simulated photon beam profiles.

Conclusions

The proposed framework is a readily applicable and customizable tool which may be applied in tuning virtual primary electron beams of Monte Carlo simulators of linear accelerators. The codes, training and test data, together with readout procedures, are freely available at the site: https://github.com/taborzbislaw/DeepBeam.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-021-01847-w.