Botnariuc, Daniela; (2024) Development and application of Monte Carlo tools in proton beam therapy. Doctoral thesis (Ph.D), UCL (University College London).

Text DB_PhDthesis.pdf - Submitted Version Access restricted to UCL open access staff until 1 January 2026. Download (66MB)

Abstract

While clinical radiotherapy pathways are well-established, it is crucial to enhance, automate, and speed up current processes to deliver treatments efficiently. This work aimed to leverage the power of Monte Carlo (MC) simulations to address key challenges in proton therapy, such as uncertainties in dose calculation algorithms used in commercial treatment planning systems (TPSs), difficulties of commissioning new proton therapy facilities and complexities of optimising 3D-printing technologies for the development of tissue-equivalent materials. First, a MC model of the pencil beam scanning (PBS) system at the University College London Hospitals (UCLH) was developed in GATE, benchmarked against experimental data and its performance was compared against that of two commercial TPSs, Eclipse and RayStation. The three models showed good agreement against the commissioning data, and dose differences of 3.5% were found for spread-out Bragg peak (SOBP) plans. Additionally, the GATE model was validated out-of-field against measurements of neutron ambient dose and maximum differences within 60% were found. To demonstrate that MC tools can support the commissioning of new proton therapy facilities and help reduce the extensive experimental measurements during the TPS dose verification, dose calculations from two TPSs were compared against measurements or GATE for 101 fields of varying complexity. It was shown that 72% and 60% of the fields were within the GATE predicted dose difference for both TPSs, for homogeneous and patient plan types, respectively, concluding that a thoroughly validated MC model can assist with the validation of the TPS during the commissioning of a new proton facility. In order to understand how clinical patient dose records can be enhanced for epidemiological studies of radiation-induced second cancers, full-body dose distributions from MC simulations and analytical models were compared. This comparison was conducted for a cohort of 20 patients who underwent abdominal irradiations, and the impact of the different dose calculation approaches on lifetime attributable risk (LAR) was evaluated. Absolute differences in LAR for the different dose calculation methods were small (within 0.24%), however relative differences in cumulative LAR were within 19%. This indicated that, for the specific cohort and analytical models applied, using full-body dosimetry obtained through analytical methods is a reasonable approximation in case MC resources are limited. Finally, the GATE model was used to develop a workflow to study how the selection of 3D-printing settings impacts the dose deposition within 3D-printed materials. It was shown that different structural configurations of the printed materials results in differences in depth-dose distributions, in terms of proton range, Bragg peak with and peak-to-plateau ratios. After an experimental validation, this framework could be used to inform upon the optimal 3D-printing parameters for the development of tissue-equivalent materials.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item