Development and Benchmarking of a Monte Carlo Dose Engine for Proton Radiation Therapy

Abstract

Dose calculation algorithms based on Monte Carlo (MC) simulations play a crucial role in radiotherapy. Here, the development and benchmarking of a novel MC dose engine, MonteRay, is presented for proton therapy aiming to support clinical activity at the Heidelberg Ion Beam Therapy center (HIT) and the development of MRI (magnetic resonance imaging)-guided particle therapy. Comparisons against dosimetric data and gold standard MC FLUKA calculations at different levels of complexity, ranging from single pencil beams in water to patient plans, showed high levels of agreement, validating the physical approach implemented in the dose engine. Additionally, MonteRay has been found to match satisfactorily to FLUKA dose predictions in magnetic fields both in homogeneous and heterogeneous scenarios advocating its use for future MRI-guided proton therapy applications. Benchmarked on 150 MeV protons transported on a 2 × 2 × 2 mm3 grid, MonteRay achieved a high computational throughput and was able to simulate the histories of more than 30,000 primary protons per second on a single CPU core.

Development and benchmarking of the first fast Monte Carlo engine for helium ion beam dose calculation: MonteRay

“…In particle therapy, inelastic nuclear scattering events generate the mixed radiation field, that is, photons, protons, neutrons, deuterons, tritons, 3 He particles, helium ions ( 4 He) and heavier fragments (nuclear recoils). For the handling of proton beams, photons and neutrons were assumed to be dosimetrically irrelevant and they were neither transported nor produced 16 . To verify that the same could be applied for helium ion beams, FLUKA simulations were run for an energy range of 50−220 MeV/u, and it was found that the contribution of photons and neutrons to the integrated depth dose distribution in water was smaller than 0.03% and therefore they are also neglected for helium ion beams.…”

“…For the handling of proton beams, photons and neutrons were assumed to be dosimetrically irrelevant and they were neither transported nor produced. 16 To verify that the same could be applied for helium ion beams, FLUKA simulations were run for an energy range of 50−220 MeV/u, and it was found that the contribution of photons and neutrons to the integrated depth dose distribution in water was smaller than 0.03% and therefore they are also neglected for helium ion beams. Like the elastic cross section, the total inelastic cross sections of helium ions are tabulated from 0.1 MeV/u to 500.1 MeV/u, however they are tabulated for all ten elements inside MonteRay, and not just for hydrogen.…”

“…In this work, we present the development and validation of MonteRay, the first rapid MC dose engine for helium ions, suitable for the energies available at HIT, covering an energy range from 0.1 MeV/u to 230 MeV/u. MonteRay provides dose predictions for proton 16 and helium ion therapy within an adaptable in‐house framework. In detail, we outline the implemented physics models for electromagnetic (energy loss and scattering) and nuclear (elastic and inelastic) interactions.…”

Magnetic resonance imaging (MRI) guided proton therapy: A review of the clinical challenges, potential benefits and pathway to implementation

“…These effects have been studied in some detail for a range of field strengths and configurations [89,[92][93][94]. Dedicated dose calculation algorithms for proton therapy have been proposed, which can enable substantial speed ups in dose calculation time without compromising accuracy [95]. The effect of magnetic fields on secondary electrons is smaller than in MRI-Linac therapy, as the energy of secondary electrons created by particle beams is lower [89,92] whilst the perturbation of proton trajectories in the patient is deterministic and can be handled with inverse planning.…”

PyMedPhys: A community effort to develop an open, Python-based standard library for medical physics applications

“…Many of these tools are in use clinically at affiliated sites, and additionally, aspects of PyMedPhys are implemented around the world for some applications. Many parties have embraced the gamma analysis module (Castle et al, 2022;Cronholm et al, 2020;Gajewski et al, 2021;Galić et al, 2020;Lysakovski et al, 2021;Milan et al, 2019;Pastor-Serrano & Perkó, 2021;Rodrıǵuez et al, 2020;Spezialetti et al, 2021;Tsuneda et al, 2021;Yang et al, 2022), while implementations of the electron cutout factor module and others (Baltz & Kirsner, 2021;Douglass & Keal, 2021;Rembish, 2021) have also been reported. Additionally, the work has been recognized by the European Society for Radiotherapy and Oncology (ESTRO) and referenced as recommended literature in their 3rd Edition of Core Curriculum for Medical Physics Experts in Radiotherapy (Bert et al, 2021).…”