Series
Doctor of Philosophy with a Major in Nuclear Engineering
Doctor of Philosophy with a Major in Nuclear Engineering
Permanent Link
Series Type
Degree Series
Description
Associated Organization(s)
Associated Organization(s)
Organizational Unit
232 results
Publication Search Results
Now showing
1 - 10 of 232
-
ItemUncertainty Analysis in ICRP 66 Human Respiratory Tract Model for Consequence Management Data Products(Georgia Institute of Technology, 2024-12-08) Margot, Dmitri EdwardInhaled radioactive is a unique hazard. Once inhaled the radioactive material is translocated within the body via incorporation into the metabolism and immune response. While metabolizing, the radioactive material is irradiating nearby tissue. Since the distribution of radioactive material changes over time, biokinetic modelling tracks the movement of the radioactive material within organs and tissues. To determine the impact of the input parameters into biokinetic modelling, a software called REDCAL (Radiation Exposure Dose Calculator) was developed in Python to handle statistically sampled parameters to compute the radiation dose from radionuclides of concern for emergency response planners. REDCAL handles the inhalation of radioactive particles and subsequent deposition within the airways. Following the deposition computations, REDCAL tracks the movement of radioactive material within the body and computes the effective dose to the individual over a lifetime. With statistically sampled input parameters, REDCAL was used to generate 3,410,000 effective dose coefficients to analyze the influence of the input parameters on the resulting dose. As sets of dose coefficients were made for each radionuclide and its associated lung clearance type(s), a defined distribution of its effective dose coefficient as a function of inhaled particle size, in AMAD, were generated to inform the sampling needing for computing derived response levels (DRLs) by in Turbo FRMAC by the Federal Radiological Monitoring and Assessment Center (FRMAC). This dissertation covers the methods, mathematics, and concepts required to compute particle deposition in the airways, solve biokinetic models, and compute effective dose from radiation sources with time-dependent concentrations.
-
ItemDevelopment of a Minibeam Delivery System for Use with a Radiation Therapy Research Platform(Georgia Institute of Technology, 2024-12-03) Herchko, StevenSpatially fractionated radiation therapy (SFRT) utilizes multiple radiation beams to combine areas of low and high dose within the treatment volume. Organ at risk (OAR) tolerance with SFRT is greater than traditional techniques, with increasing benefits as the beam size decreases. Minibeam radiation therapy (MBRT) is a form of SFRT with beam sizes on the order of 1 mm, which is achievable utilizing conventional treatment techniques, but MBRT has only recently been used in the treatment of humans. The primary objectives of this work are to develop a system to deliver minibeam radiation therapy (MBRT) using a commercially available small animal research radiation therapy platform, perform cell survival studies comparing the survival of cells irradiated with both MBRT and broad beam (BB) dose delivery, and complete an in vivo treatment planning study to determine the dosimetric differences of different MBRT treatment delivery techniques. Most work in SFRT has focused on larger beams that provide limited benefits or very small beams that are too impractical for clinical use. SFRT is an area of active research with growing clinical applications and interest. This work aims to lay the groundwork for future small animal studies and inform the development of future clinical trials in humans.
-
ItemAI-Driven 4D Lung SBRT Treatment Planning: An End-to-End Approach(Georgia Institute of Technology, 2024-09-04) Matkovic, Luke AlexanderCurrently, patient motion management in 4D lung SBRT cases is primarily based on subjective judgment from the involved staff, namely the radiation oncologist. To obtain more objective information without adding significant cost overhead to the clinical workflow, this work improves registration and segmentation methods using deep learning techniques and utilizes built-in treatment planning software in Eclipse. Image registration and OAR segmentation are time-consuming and error-prone steps in the treatment planning workflow for external beam radiation. In this work, a registration method is developed to improve numerical results and deformation realism. The registration model uses cycle consistency to improve bi-directional registration consistency, a generator-discriminator pair to force realistic deformations, and weak supervision to improve registration quality. Additionally, segmentation models were created and compared to current industry-standard techniques and radiation oncologist-approved contours using Eclipse. Eclipse is further used to generate DVH curves to aid radiation oncologist decision making regarding motion management selection for 4DCT lung SBRT cases. These steps can be utilized in a single, mostly automated workflow to aid in phase gating selection.
-
ItemDesign and optimization of mini-beam GRID and LATTICE therapy using linear accelerators(Georgia Institute of Technology, 2024-07-19) Carter, Zachary GrayDose-Volume effects have been shown to increase the radiation dose tolerance of critical organs-at-risk (OARs) in multiple normal tissues such as the brainstem, spinal cord, parotid glands and in animal models. The irradiated volume must be kept “small” on the order of millimeters for tissues to experience additional protection from dose-volume effects. Mouse ears have demonstrated minimal toxicity from more than 60 Gy of single-fraction dose when the beam width was kept below 2 mm. Spatially Fractionated Radiation Therapy (SFRT) aims to deliver heterogeneous peak and valley dose regions to the tumor in contrast to homogeneous dose techniques such as intensity modulated radiotherapy (IMRT). A form of SFRT called minibeam radiation therapy (MBRT) typically uses beams on the order of sub-millimeter to a few millimeters which is within the size range for exploitation of dose-volume effects, but the connection between SFRT, MBRT and dose-volume effects has not been thoroughly explored. Glioma and glioblastoma are aggressive, radioresistant primary brain tumors with bleak prognosis and minimal treatment options, but LINAC-based photon MBRT in an animal study demonstrated complete tumor response for glioma when compared to stereotactic radiosurgery which only showed partial tumor remission. These results motivated this thesis work to develop novel techniques for delivering LINAC-based MBRT for overcoming radioresistance of aggressive tumors such as glioma/glioblastoma. In this thesis, the novel techniques of 2D “en face” honeycomb LINAC-MBRT, 3D “mini-LATTICE”, and “4π-MBRT” are developed and explained. This thesis will show that with modifications to existing commercially available medical linear accelerators, LINAC-MBRT can be implemented with MV-photon beams to increase the therapeutic ratio and treatment outcomes of radioresistant diseases such as glioma/glioblastoma.
-
ItemNovel Techniques for Cell-by-Cell Monte Carlo Simulation of Radiation-Induced DNA Damage and Repair with Models of Specific Chromosome Structures(Georgia Institute of Technology, 2024-04-27) Andriotty, MatthewNew tools have recently been developed to simulate nano-scale radiation track structure and chemical species resulting from the radiolysis of water, record the resulting DNA damage, and model the pathways of DNA repair. These tools allow for in silico investigation of cell radiosensitivity and how it is influenced by several factors, such as the arrangement of chromosomes in the cell nucleus. (1) This work presents a pipeline employing in vitro Hi-C chromosome conformation capture data to create specific cell nucleus models which are then used in TOPAS-nBio radiation simulations. The DSBs are recorded, and then their repair is mechanistically simulated in MEDRAS-MC. This pipeline is used to predict not only the initial radiation-induced DNA damage, but also the repair outcomes resulting from this damage in order to investigate the role chromosome conformation plays in the biological outcome of radiation exposure. (2) This work also establishes a method using libraries of DNA damage files, each consisting of DSB data resulting from a single track of high-LET radiation, to efficiently predict DNA repair outcomes in a variety of scenarios. These libraries are created by running many TOPAS-nBio simulations of a single proton track for a range of energies and recording the DNA damage. A suitable number of these data files are superimposed according to the desired dose, type, and energy of radiation. Then, MEDRAS-MC is used to simulate the repair outcomes of this superimposed radiation damage. These DNA damage data libraries allow for cell-by-cell radiobiological simulation of multiple applications, such as proton or neutron therapy, without the need for time-consuming brute force TOPAS-nBio simulations of high-LET radiation track structure. (3) Additionally, this work demonstrates the incorporation of DNA damage and repair simulations into a larger multiscale framework to simulate the effects of radiation from the whole-body scale to the cellular scale. DNA damage and repair simulations are conducted for individual cells in a multicellular tumor model, and the results are used to calculate linear-quadratic cell-survival curves for the tumor's cells.
-
ItemDevelopment of Coupled Machine Learning and Optimization Framework for Comprehensive Molten Salt Reactor Design(Georgia Institute of Technology, 2023-12-18) Kazaroff, Coral HannahThe continued growth of the world population and push for clean energy has created a political impetus to modernize the existing nuclear reactor fleet. Such efforts have culminated in the creation of the Generation IV International Forum, a multinational coalition formed with the purpose of researching and testing six advanced reactor prototypes. Of these designs, the molten salt reactor (MSR) has received notable attention due to its attractive characteristics of high thermodynamic efficiency, inherent safety features, and improved proliferation resistance. Despite these benefits, significant work remains to be done before MSRs can be deployed to the commercial grid. The variation in designs poses a significant challenge towards down selection and necessitates optimization methods that can reliably provide information about variables associated with different design metrics. This work addresses this challenge through a comprehensive modeling framework developed for both moderated and unmoderated liquid-fueled molten salt reactors. It does so through an extensive thermal hydraulic/neutronic coupled, time-dependent database of reactor designs that is dependent on multiple input variables associated with geometry, materials, and fuel salts. From this database, a predictive machine learning model is trained to determine output parameters given any set of input reactor design variables. This model is then used in a genetic algorithm optimization sequence. The optimization sequence allows for flexibility in the optimization process, where different metrics can be chosen for different design goals. For example, one case could involve multiplication factor constraints coupled with minimization of transuranics production, whereas another could focus on maximizing breeding ratio instead. The end result of the framework provides a potential core designer with the required input metrics for building a molten salt reactor based on user-defined performance metrics.
-
ItemA Hybrid Radiation Transport Detector Response Function Methodology for Modeling Contaminated Sites(Georgia Institute of Technology, 2023-10-27) Asano, Ethan AkiraTo date, guidance for environmental assessment using experimental radiation detection techniques – specifically relating photon detector responses to dose or cancer risk-based radionuclides concentrations in environmental media – exist under the Environmental Protection Agency (EPA) Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). In recent years, the EPA has undertaken efforts to develop the “U.S. EPA Superfund Counts-Per-Minute (CPM) Electronic Calculator” web-based field screening tool to address this technical challenge. The CPM Calculator predicts photon detector readings in CPM that, if exceeded in the field, may warrant further action (e.g., cleanup) when demonstrating regulatory compliance in the radiological site remediation process. A modeling research effort was undertaken to estimate thousands of sodium iodide thallium-doped (NaI(Tl)) detector total efficiency responses in the development of a methodology to calculate detector response from simulated contamination in various geometries in environmental media. A detector response function (DRF) methodology was developed by coupling capabilities of the stochastic Monte Carlo (MC) method with the Consistent Adjoint Driven Importance Sampling (CADIS) hybrid radiation transport method to estimate the NaI(Tl) detector efficiency responses for a broad range of wide-area contamination scenarios. The CADIS hybrid method, available through the massively parallel MC code Shift, was harnessed to significantly increase the computational efficiency of thousands of Monte Carlo N-Particle (MCNP) pulse height tally (PHT) simulations involving wide-area, distributed photon source contamination in environmental media (surface source contamination scenarios were evaluated using only MCNP). Total detector efficiency responses were estimated for a breadth of contamination scenarios based on the following parameters: discrete monoenergetic source energy/radionuclide spectrum, contaminated medium type, depth of contamination, modeled cylindrical NaI(Tl) detector configuration, and source-detector distance. Simulated results were post-processed to yield calibration factors in units of CPM per surface or distributed source activity concentration (i.e., CPM/(pCi/〖cm〗^2 ) or CPM/(pCi/g), respectively). At the time of writing, all calibration factor data has been implemented into the CPM Calculator – the final version is expected to be released following physical laboratory validation in partnership with EPA laboratories. In practice, field surveying personnel will be able to quickly compare a measured NaI(Tl) detector net count rate to a threshold value derived from a predetermined dose or risk limit (i.e., calibration factor normalized by an activity concentration derived from a dose or risk limit) for a variety of wide-area contamination scenarios. Ultimately, currently expensive and time-consuming sample collection and laboratory analyses efforts required for demonstrating regulatory compliance in future wide-area contamination assay applications may be reduced. This work results from a comprehensive study that consists of multiple auxiliary efforts following the development of the DRF methodology described previously, which we refer to as “Shift with CADIS.” A study was conducted to analyze the effectiveness of Shift with CADIS compared to various radiation transport variance reduction (VR) methods for improved modeling in future wide-area contamination assay applications. The latter is intended for publication outside the current study and is therefore only adverted herein. Instead, a follow-up analysis that further explores alternative methods for approximating solutions to the wide-area contamination transport problems of interest is presented. Additionally, a method was developed for estimating the total detector efficiency response contribution from secondary photons arising from electron source emissions (called bremsstrahlung) in soil with MC simulation, which employed the electron transport condensed history method in a computationally feasible manner. Lastly, investigatory measurements were performed with laboratory-scale setups of low-level activity environmental soil contamination with field surrogate photon source terms to validate simulated detector efficiency responses.
-
ItemOptimization of Beamline Elements and Shielding in a Preclinical MV Bremsstrahlung FLASH Irradiator(Georgia Institute of Technology, 2023-07-25) Rosenstrom, Andrew RyanFLASH radiotherapy is an emerging modality that takes advantage of observed tissue-sparing effects that occur at dose rates above 40 Gy/s. While the so-called FLASH Effect has been shown to occur using electrons, protons, and low-energy photons (<600 keV), the underlying biological mechanism is still disputed. In order to obtain greater clarity regarding the biological mechanism, preclinical, experimental systems must be created with irradiation parameters that span a wide range of achievable dose rates and pulse frequencies. The FLASH-Experimental X-ray small Animal Conformal Therapy system, as a part of the Pluridirectional High-energy Agile Scanning Electronic Radiotherapy project, is a preclinical external beam radiotherapy device that uses high-energy bremsstrahlung (>10 MeV) using a high-power electron beam (>12 kW) produced by a novel compact linear accelerating structure built at SLAC National Accelerator Laboratory expanding the range of photon energies that have investigated the FLASH effect. Due to the high-energy bremsstrahlung radiation and high workload, shielding and beam-shaping solutions are needed to create a safe and well-characterized experimental apparatus to raise the level of technological readiness in anticipation of clinical FLASH radiotherapy machines. This work focuses on the design and metaheuristic optimization of the beamline elements and shielding, as well as the experimental verification of methods used in their development. The novel application of multilayered shielding produces volumetric and mass-efficient shielding that mitigates photoneutron contamination and reduces the shielding burden on the radiation vault used to house the FLASH-EXACT. A comparison is made between common radiation shielding assessment tools used in MeV bremsstrahlung radiotherapy machines: Monte Carlo radiation transport using the code FLUKA, and empirical methods reported in the National Council on Radiation Protection and Measurement Report 151. The effects of the assumptions used in the formulation of the empirical method are shown to lead to an overly conservative estimation of the dose to personnel when polyethylene is incorporated in the treatment head shielding. The beamline elements that shape the transverse and axial dose distribution in the experimental sample are optimized through a flexible hybridized Genetic and Nelder-Mead Simplex Search Algorithm, which automates the production of high-performing configurations of the beamline for a variety of desired field sizes (0.1225 cm2 at 56 Gy/s and 1 cm2 at 435 Gy/s) while simultaneously incorporating thermal protection and electron contamination criteria for the bremsstrahlung converter. To further address the thermal challenges associated with bremsstrahlung-based FLASH radiotherapy machines, design sensitivities of the vacuum window and new bremsstrahlung converter configurations are assessed using two finite element analysis codes to ensure there is a reasonable safety margin with respect to thermal and structural loading failure during the operation of the FLASH-EXACT system.
-
ItemCharacterization and Validation of a Proton Minibeam System and Evaluation of Treatment Planning Systems(Georgia Institute of Technology, 2023-05-16) Stanforth, AlexanderSpatially fractionated radiation therapy (SFRT) has been used to deliver radiation therapy since the early 1900s, with renewed interest in the last few decades. Proton SFRT is a recent proposal which is used in limited clinical circumstances. Further research on proton mini- and microbeams have been shown to have an increased therapeutic index when compared to traditional SFRT. Despite the body of research showing its biologic benefits compared to other more traditional methods of radiation therapy, proton minibeam radiation therapy (pMBRT) is not currently used clinically. This work attempts to model a pMBRT minibeam and assess the accuracy of Class-II Monte Carlo algorithms in generating clinical treatment plans. This work characterized a collimated proton minibeam system using TOPAS simulations and experimental measurements. It then used the open-source treatment planning system (TPS), MCSquare, to compare the results of Class-II Monte Carlo methods to more extensive Monte Carlo simulations and measurements. A simple clinical case with an anthropomorphic phantom was used to test MCSquare end-to-end. This work further characterized the secondary neutral particles present in a collimated minibeam system as well as the impact the collimator has on linear energy transfer (LET). Overall, measurements showed high agreement with TOPAS simulations when characterizing the minibeam. The TPS results showed high agreement with TOPAS and measurements, with noted differences due to lack of secondary neutral particle transport. The end-to-end test showed high agreement with a gamma analysis. The secondary neutral particles and LET characterization showed high agreement between measurements and TOPAS simulations and the impact of the collimator on these quantities wasstudied.
-
ItemTime Dependent Full Core Coupled Multiphysics Analysis of Nuclear Thermal Propulsion Reactors(Georgia Institute of Technology, 2023-04-25) Krecicki, Matthew A.A novel full-core multiphysics analysis framework for Nuclear Thermal Propulsion (NTP) Reactors is developed in this dissertation. To achieve the high specific impulses and thrust levels required for crewed space exploration missions NTP systems operate at very high temperatures, rely on complex counter-flow needed to drive the turbopump, and exhibit dynamic behavior for short pulse-like operation. Therefore, the design and analysis of a NTP reactor-core requires multiphysics computational tools that can capture the heat transfer and flow complexities and dynamic haviour during the engine operation. Existing higher-order codes, such as ANSYS, can analyze complex flow paths in a NTP reactor, but incur prohibitively large computational costs and are not applicable for full-core multiphysics analysis. The development and verification of the reduced-order, ntpThermo, code is a novel contribution as it is capable of accurately modeling the complex flow paths and heat transfer within an NTP reactor. In addition, ntpThermo can perform coupled thermal-hydraulic thermo-mechanical analysis to capture the impact of thermal expansion with an acceptable computational cost. The ntpThermo code is coupled to the Monte Carlo Neutron transport Serpent code via the novel Basilisk multiphysics framework. The Basilisk framework enables full-core time-dependent multiphysics analysis by leveraging the pre-existing depletion solvers implemented into the Serpent code. The framework also enables the user to perform a critical drum search during each depletion step to account for the impact of control drum rotation during operation. Previous NTP-related research that focused on full core design has applied decoupled analysis approaches where the impact of thermal-hydraulic and thermo-mechanical feedback on the neutronic solution is neglected. In an effort to provide useful insights for current programs a reactor design which adheres to the current industry ground rules was developed. The subsequent analysis demonstrates that such decoupled approaches can introduce significant errors in the spatial power distributions and thus predicted thermal and mechanical safety margins. More specifically, for heavily moderated High Assay-Low Enriched Uranium fueled designs the fuel and moderator temperature spatial distributions have a significant impact on the neutron economy and spatial power distributions. Additionally, the impact of thermo-mechanical feedback has a significant impact on the mass-flow distribution within the core, and thus the solid material temperatures. Due to the elevated exit gas temperatures required to satisfy rocket engine performance requirements orificing is typically applied to the fuel elements in the core to reduce peak fuel temperatures. When a consistent multiphysics design approach is applied to design the orificing pattern a constant peak fuel temperature can be maintained through a 60-minute full-power burn due to the balance of various multiphysics feedback mechanisms. This dissertation demonstrates the importance of multiphysics tools to design a NTP reactor that can maintain adequate thermal and mechanical safety margins while also satisfying engine performance requirements.