Sendo o método mais preciso para estimar a dose absorvida em radioterapia, o Método de Monte Carlo (MMC) tem sido amplamente utilizado no planejamento de tratamento radioterápico. No entanto, a sua eciência pode ser melhorada para aplicações clínicas de rotina. Nesta dissertação é apresentado o código CUBMC, um código de Monte Carlo que simula o transporte de fótons para cálculo de dose, desenvolvido na plataforma CUDA (Compute Unified Device Architecture). A simulação de eventos físicos é baseada no algoritmo presente no código PENELOPE, e as tabelas de seção de choque utilizadas são geradas pela rotina MATERIAL, também presente no código PENELOPE. Os fótons são transportados em objetos simuladores descritos por voxels. Existem duas abordagens distintas utilizadas para a simulação. A primeira delas obriga o fóton a realizar uma parada toda vez que cruza a fronteira de um voxel, a segunda e pelo Método de Woodcock, onde o fóton ignora a existência de fronteiras e é transportado em um meio homogêneo fictício. O código CUBMC tem como objetivo ser uma opção de código simulador que, ao utilizar a capacidade de processamento paralelo de unidades de processamento gráfico (GPU), apresente alto desempenho em máquinas compactas e de baixo custo, podendo assim ser aplicado em casos clínicos e incorporado a sistemas de planejamento de tratamento em radioterapia. / As the most accurate method to estimate absorbed dose in radiotherapy, Monte Carlo Method (MCM) has been widely used in radiotherapy treatment planning. Nevertheless, its efficiency can be improved for clinical routine applications. In this master thesis, the CUBMC code is presented, a GPU-based MC photon transport algorithm for dose calculation under the Compute Unified Device Architecture (CUDA) platform. The simulation of physical events is based on the algorithm used in PENELOPE, and the cross section table used is the one generated by the MATERIAL routine, also present in PENELOPE code. Photons are transported in voxel-based geometries with different compositions. There are two distinct approaches used for transport simulation. The first of them forces the photon to stop at every voxel frontier, the second one is the Woodcock method, where the photon ignores the existence of borders and travels in homogeneous fictitious medium. The CUBMC code aims to be an alternative for Monte Carlo simulator code that, by using the capability of parallel processing of graphics processing units (GPU), provides high performance simulations in low cost compact machines, and thus can be applied in clinical cases and incorporated in treatment planning systems for radiotherapy.

Criticality calculations are often performed in MCNP5 using the Shannon entropy as an indicator of source convergence for the given neutron transport problem. The Shannon entropy is a concept that comes from information theory. The Shannon entropy is calculated for each batch in MCNP5, and it has been shown that the Shannon entropy tends to converge to a single value as the source distribution converges. MCNP5 has its own criteria for when the Shannon entropy has converged and recommends a number for how many batches should be skipped; however, this value for how many batches should be skipped is often not very accurate and has room for improvement. This work will investigate an approach for using the Shannon entropy source distribution convergence information obtained in a shorter simulation to predict the required number of generations skipped in the reference case with desired statistical precision. In several test cases, it has been found that running a lesser number of particles per batch produces a similar Shannon entropy graph when compared to running more particles per batch. Then, by appropriate adjustment through a synthetic model, one is able to determine when the Shannon entropy will converge by running fewer particles, finding the point where it converges and then using this value to determine how many batches one should skip for a given problem. This reduces computational time and any "guessing" involved when deciding how many batches to skip. Thus, the purpose of this research is to develop a model showing how one can use this concept and produce a streamlined approach for applying this concept to a criticality problem.

Counting efficiencies were determined by empirical measurement and Monte Carlo simulation for carbon-14, strontium-89, strontium-90, and yttrium-90 standards counted by low-background gas flow proportional counter for strontium carbonate precipitates in the range from 3 to 33 mg cm⁻². The maximum beta particle energies range from 0.156 MeV for carbon-14 to 2.28 MeV for yttrium-90. The parameters for estimating the counting efficiency are summarized for sources with areal thickness of 14 mg cm⁻² and over the range in strontium carbonate areal thickness from 0.1 mg cm⁻² to 33 mg cm⁻². Uncertainty budgets providing estimates of the uncertainty, sources of variability in the calibration process, and the total expanded uncertainty are presented. Information is presented for the Monte Carlo simulation regarding the composition of the detector window, the energy excluded by the amplifier discriminator of the counting system, and the physical density of materials for this analytical process. The histogram normalization routine implemented within MCNP is described and found to bias the probability distribution for beta-particle energy spectra. The difference in the specification of the probability distribution for beta-particle energy spectra in ICRU 56 Appendix D and MCNP requirements are described and a correction for the bias introduced during the normalization process for beta spectra is provided. Counting efficiencies determined by empirical measurement and Monte Carlo simulations agree within the total expanded uncertainties of the measurements and the uncertainties of the Monte Carlo simulations.

Irradiation experiments for the pebble bed modular reactor PBMR fuel (coated fuel particles and pebble fuel) are planned at the South African First Atomic Reactor Installation (SAFARI-1). The experiments are conducted to investigate the behavior of the fuel under normal operating and accelerated/accident simulating conditions because the safe operation of the reactor relies on the integrity of the fuel for retention of radioactivity. For fuel irradiation experiments, the accurate knowledge and analysis of the neutron spectrum of the irradiation facility is required. In addition to knowledge of the neutron spectrum in the irradiation facility, power distributions and knowledge of nuclear heating values has to be acquired. The SAFARI-1 reactor boosts operating fluid temperatures of about 300 K. On the contrary, the PBMR can reach temperatures in up to about 1370 K under normal operating conditions. This calls for design of high temperature irradiation rigs for irradiation of the PBMR fuel in the SAFARI-1 reactor. The design of this instrument (rig) should be such that to create an isolated high temperature environment in the SAFARI-1 reactor, to achieve the requirements of the PBMR fuel irradiation program. The design of the irradiation rig is planned such that the rig should fit in the existing irradiation channels of the SAFARI-1 reactor, a time and cost saving from the licensing perspective. This study aims to establish the know-how of coated particle and pebble modeling in using the Monte Carlo N-Particle code (MCNP5). The study also aims to establish the know-how of rig design. In this study, the Necsa in-house code Overall System for the Calculation of Reactors (OSCAR-3), a software known as OScar 3-Mcnp INTerface (OSMINT) linking OSCAR-3 and MCNP5, also developed at Necsa, as well as MCNP5 code developed and maintained by the Los Alamos team, are used to calculate neutronic and power distribution parameters that are important for fuel irradiations and for rig design. This study presents results and data that can be used to make improvements in the design of the rig or to confirm if the required operational conditions can be met with the current preliminary rig design. Result of the neutronic analysis are presented for the SAFARI-1 core, core irradiation channel B6 (where the PBMR fuel irradiation rig is loaded for the purpose of this study), the rig structure and the pebble fuel are presented. Furthermore results of the power distribution and nuclear heating values in the reactor core, the irradiation channel B6, the rig structures and the pebble fuel is also presented. The loading of the PBMR fuel irradiation rig in core position B6 reduces the core reactivity due to the fact that the loading of the rig displaces the water moderator in channel B6 introducing vast amounts of helium. This impacts on the keff value because there will be less neutron thermalization and reproduction due to the decreased population of thermal neutrons. The rig is found to introduce a negative reactivity insertion of 46 pcm. The loading of this rig in the core leads to no significant perturbations on the core power distribution. The core hottest channel is still localized in core channel C6 both with RIG IN and RIG OUT cases. A power tilt is observed, with the south side of the core experiencing reduced assembly averaged fission power, with correspondingly small compensations from the assemblies on the north side of the core. The perturbations on the core assembly averaged fluxes are more pronounced in the eight assemblies surrounding B6. Core position B6 suffers an 18% neutron flux depression with the loading of the rig. The fluxes in core positions A5, A6, A7, B5, B7 and C7 are increased when the rig is loading. The largest increases are noted as 12% in A7, 9% in A6 and 6% in A5 and B7. All the eight core positions surrounding B6 experience reduced photon fluxes with the loading of the rig. Core position B6 shows a flux depression of up to 20%, with 10% reduction in core position A6. The remainder seven positions surrounding B6 shows flux depressions of no more than 5%. Further on, due to decreased moderation effects, the axial neutron flux in core position B6 is reduced by 20% when the rig is loaded. The energy dependent neutron flux in B6 decreases by 50% in the thermal energy range with corresponding increases of up to 50% in the resonance and fast energy regions. The axial and the energy dependent photon flux in core position B6 decreases by up to 20% when the rig is loaded. The magnitude of the neutron and photon fluxes is found to have a direct proportion on the neutron and photon heating values. While the amount of neutron heating in core position B6 increases by one order of magnitude, when the rig is loaded, the photon heating values increases by up to 60% in the region spanning ±10cm about the core centerline. The amount of photon heating in the rig structural materials dominates neutron heating, except in the helium regions of the rig, where neutron heating dominates photon heating. In the fuel region of the pebble, fission heating (3803W) largely dominates photon heating (119W). / Thesis (M.Sc. (Nuclear Engineering))--North-West University, Potchefstroom Campus, 2009

Irradiation experiments for the pebble bed modular reactor PBMR fuel (coated fuel particles and pebble fuel) are planned at the South African First Atomic Reactor Installation (SAFARI-1). The experiments are conducted to investigate the behavior of the fuel under normal operating and accelerated/accident simulating conditions because the safe operation of the reactor relies on the integrity of the fuel for retention of radioactivity. For fuel irradiation experiments, the accurate knowledge and analysis of the neutron spectrum of the irradiation facility is required. In addition to knowledge of the neutron spectrum in the irradiation facility, power distributions and knowledge of nuclear heating values has to be acquired. The SAFARI-1 reactor boosts operating fluid temperatures of about 300 K. On the contrary, the PBMR can reach temperatures in up to about 1370 K under normal operating conditions. This calls for design of high temperature irradiation rigs for irradiation of the PBMR fuel in the SAFARI-1 reactor. The design of this instrument (rig) should be such that to create an isolated high temperature environment in the SAFARI-1 reactor, to achieve the requirements of the PBMR fuel irradiation program. The design of the irradiation rig is planned such that the rig should fit in the existing irradiation channels of the SAFARI-1 reactor, a time and cost saving from the licensing perspective. This study aims to establish the know-how of coated particle and pebble modeling in using the Monte Carlo N-Particle code (MCNP5). The study also aims to establish the know-how of rig design. In this study, the Necsa in-house code Overall System for the Calculation of Reactors (OSCAR-3), a software known as OScar 3-Mcnp INTerface (OSMINT) linking OSCAR-3 and MCNP5, also developed at Necsa, as well as MCNP5 code developed and maintained by the Los Alamos team, are used to calculate neutronic and power distribution parameters that are important for fuel irradiations and for rig design. This study presents results and data that can be used to make improvements in the design of the rig or to confirm if the required operational conditions can be met with the current preliminary rig design. Result of the neutronic analysis are presented for the SAFARI-1 core, core irradiation channel B6 (where the PBMR fuel irradiation rig is loaded for the purpose of this study), the rig structure and the pebble fuel are presented. Furthermore results of the power distribution and nuclear heating values in the reactor core, the irradiation channel B6, the rig structures and the pebble fuel is also presented. The loading of the PBMR fuel irradiation rig in core position B6 reduces the core reactivity due to the fact that the loading of the rig displaces the water moderator in channel B6 introducing vast amounts of helium. This impacts on the keff value because there will be less neutron thermalization and reproduction due to the decreased population of thermal neutrons. The rig is found to introduce a negative reactivity insertion of 46 pcm. The loading of this rig in the core leads to no significant perturbations on the core power distribution. The core hottest channel is still localized in core channel C6 both with RIG IN and RIG OUT cases. A power tilt is observed, with the south side of the core experiencing reduced assembly averaged fission power, with correspondingly small compensations from the assemblies on the north side of the core. The perturbations on the core assembly averaged fluxes are more pronounced in the eight assemblies surrounding B6. Core position B6 suffers an 18% neutron flux depression with the loading of the rig. The fluxes in core positions A5, A6, A7, B5, B7 and C7 are increased when the rig is loading. The largest increases are noted as 12% in A7, 9% in A6 and 6% in A5 and B7. All the eight core positions surrounding B6 experience reduced photon fluxes with the loading of the rig. Core position B6 shows a flux depression of up to 20%, with 10% reduction in core position A6. The remainder seven positions surrounding B6 shows flux depressions of no more than 5%. Further on, due to decreased moderation effects, the axial neutron flux in core position B6 is reduced by 20% when the rig is loaded. The energy dependent neutron flux in B6 decreases by 50% in the thermal energy range with corresponding increases of up to 50% in the resonance and fast energy regions. The axial and the energy dependent photon flux in core position B6 decreases by up to 20% when the rig is loaded. The magnitude of the neutron and photon fluxes is found to have a direct proportion on the neutron and photon heating values. While the amount of neutron heating in core position B6 increases by one order of magnitude, when the rig is loaded, the photon heating values increases by up to 60% in the region spanning ±10cm about the core centerline. The amount of photon heating in the rig structural materials dominates neutron heating, except in the helium regions of the rig, where neutron heating dominates photon heating. In the fuel region of the pebble, fission heating (3803W) largely dominates photon heating (119W). / Thesis (M.Sc. (Nuclear Engineering))--North-West University, Potchefstroom Campus, 2009

La principal motivación que ha propiciado el veloz progreso de las técnicas de prevención y tratamiento del cáncer en los últimos años ha sido, y continúa siendo, su protagonismo en las listas de principales causas de muerte: más de 10 millones de diagnósticos anuales a escala global y más de 160.000 en territorio español. En este contexto, la implementación clínica de los Sistemas de Planificación de Tratamientos de Radioterapia (RTPS) ha desempeñado un papel capital. Resulta lugar común en el ámbito de la medicina nuclear que los algoritmos convencionales de cálculo de dosis que poseen los RTPS, de naturaleza determinista, carecen de la precisión necesaria a la hora de determinar el transporte lateral de electrones cuando un haz de partículas cargadas incide en la interfaz entre un medio material de densidad baja y otro de densidad alta; además, incurren en predicciones de dosis erróneas ante la presencia de heterogeneidades debido a la alta dispersión de electrones que se produce entre los distintos materiales. Se ha comprobado que los métodos de cálculo de dosis basados en Monte Carlo (MC) proporcionan distribuciones de dosis más precisas que los algoritmos convencionales en los planificadores 3D comerciales. Sin embargo, pese a la substancial mejora que ofrecen los primeros, aún no se han conseguido implementar de forma extensiva en el ámbito clínico debido al coste de tiempo computacional que requieren para obtener resultados con una estadística aceptable. Esta tesis presenta un estudio de integración de cálculos dosimétricos realizados con un código de transporte de partículas basado en Monte Carlo (MCNP) en un Sistema de Planificación de Tratamientos de distribución libre (PlanUNC), análogo a los comerciales. El trabajo comprende no sólo la consecución de un software que permite la intercomunicación de MCNP con PLUNC, al que se designa con el nombre de MCTPS-UPV, sino también un estudio de optimización de la simulación MC con objeto de agilizar el cálculo y minimizar su tiempo de computación, sin perjuicio de obtener resultados estadísticamente válidos. Los resultados demuestran que, acoplando en PLUNC el código MCNP en su versión 5 1.40 (y partiendo de la suposición de que los resultados de MCNP5 se ajustan a los experimentales en un intervalo de error del 5%, puesto que han sido validados experimentalmente en una cuba de agua con heterogeneidades con el acelerador lineal (LinAc) Elekta Precise y un colimador multiláminas (MLC)), puede efectuarse dicha simulación en pacientes reales mediante una metodología que permite tiempos computacionales aptos para su aplicación clínica y deposiciones de dosis precisas en medios heterogéneos. La investigación proporciona, además, de forma académica, un estudio extensivo tanto práctico como teórico en torno a la simulación MC en sistemas de planificación de tratamientos y a las particularidades asociadas a la implementación clínica de los algoritmos dosimétricos MC, tales como la influencia de las heterogeneidades en la deposición de dosis en el paciente, la influencia del tamaño de la voxelización o la reducción de varianza en el cálculo estadístico, tan importantes en el contexto en que ésta se inscribe. Las simulaciones se llevan a cabo mediante un LinAc Elekta Precise con MLC y distintos tamaños y conformaciones de campo que permiten un análisis exhaustivo de todas las variables que participan en la irradiación. Finalmente, el trabajo debe derivar en una futura validación experimental de las distribuciones de dosis dentro del maniquí RANDO mediante dosímetros, además de en la posibilidad de obtener tiempos de cálculo realistas mediante tecnologías más accesibles al usuario, en la posibilidad de incluir una conformación del haz posterior a la simulación incial del espacio de fase o en el estudio de la contaminación del paciente por fotoneutrones. / Abella Aranda, V. (2014). Sistema de Planificación de Tratamientos de Radioterapia para Aceleradores Lineales de Partículas (LinAc) basado en el método Monte Carlo [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/43219