PyMCGPU-IR, a computational Monte Carlo application for occupational dosimetry in interventional radiology procedures
(English) Interventional radiology procedures expose both patients and operators to radiation. It is crucial to minimise this exposure and comply with dose limits while ensuring the success of the procedure. Personal protective equipment and dosimetry are essential components in achieving these obje...
| Autor: | |
|---|---|
| Tipo de documento: | tese |
| Estado: | Versão publicada |
| Data de publicação: | 2024 |
| País: | España |
| Recursos: | CBUC, CESCA |
| Repositório: | TDR. Tesis Doctorales en Red |
| OAI Identifier: | oai:www.tdx.cat:10803/693713 |
| Acesso em linha: | http://hdl.handle.net/10803/693713 https://dx.doi.org/10.5821/dissertation-2117-424717 |
| Access Level: | Acceso aberto |
| Palavra-chave: | Àrees temàtiques de la UPC::Física 539 |
| Resumo: | (English) Interventional radiology procedures expose both patients and operators to radiation. It is crucial to minimise this exposure and comply with dose limits while ensuring the success of the procedure. Personal protective equipment and dosimetry are essential components in achieving these objectives. However, occupational dosimetry presents challenges due to inhomogeneous radiation fields, which imply that operators should wear multiple dosemeters, although typically a single dosemeter is used in a single location to measure personal dose equivalent. Computational methods offer an alternative for occupational dosimetry, allowing for dose calculation at multiple points. In particular, Monte Carlo codes have long been used for simulating the transport of radiation particles. This thesis focuses on the development and validation of PyMCGPU-IR, an application designed to estimate occupational doses in interventional radiology procedures, operating automatically without user intervention and doing so within short time frames. PyMCGPU-IR is based on the Monte Carlo code MCGPU-IR, which leverages the computational power of GPUs to simulate particle trajectories simultaneously. MCGPU-IR can calculate deposited dose values in voxelized geometries of patients and operators in less than two minutes per event. PyMCGPU-IR extracts procedural information from the RDSR file and the main operator’s position from a tracking camera system, automating the execution of the MCGPU-IR code and providing the desired dose values. The first step of this thesis was the validation of the MCGPU-IR code, comparing its occupational dose results with those of the standard Monte Carlo code PENELOPE. Subsequently, its results were compared with occupational dose values measured with dosemeters, revealing a good agreement. However, scenarios involving shielding showed a tendency of MCGPU-IR to underestimate the occupational dose. PyMCGPU-IR has been evaluated in real clinical scenarios, using phantoms that simulate the bodies of patients and operators, with satisfactory results. However, in real medical interventions, certain information required for PyMCGPU-IR, such as the patient table position, the presence of shielding, or the operator's position, was often unavailable and had to be manually entered into the application based on assumptions. As a result, PyMCGPU-IR still has limitations in fully automating the process in real medical interventions. Nevertheless, once these assumptions are considered, PyMCGPU-IR can calculate personal dose equivalent Hp(10) values in operators with differences compared to experimental measurements, within the accepted tolerance in occupational dosimetry. In addition to calculating Hp(10), PyMCGPU-IR can also calculate organ dose values, effective dose for operators, and skin dose values for patients in less than two minutes per irradiation event. Furthermore, it provides deposited dose values in voxels, facilitating graphical visualisation of dose distribution. |
|---|