Development of advanced geometric models and acceleration techniques for monte carlo simulation in medical physics
- Autores: Andreu Badal Soler
- Directores de la Tesis: Josep Sempau Roma (dir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2008
- Idioma: español
- Tribunal Calificador de la Tesis: Xavier Ortega Aramburu (presid.), Francesc Salvat (secret.), Christoph Hoeschen (voc.), Domenec Ros (voc.), Aldo Badano (voc.)
- Programa de doctorado: FISICA I ENGINYERIA NUCLEAR
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
Monte Carlo simulation of radiation transport is currently applied in a large variety of areas, However, the geometric models implemented in most general-purpose codes impose limitations on the shape of the objects that can be defined. These models are not well suited to represent the free-form (i.e., arbitrary) shapes found in anatomic structures or complex medical devices. As a result, some clinical applications that require the use of highly detailed phantoms can not be properly addressed.
This thesis is devoted to the development of advanced geometric models and acceleration techniques that facilitate the use of state-of-the-art Monte Carlo simulation in medical physics applications involving detailed anatomical phantoms. To this end, two new codes, based on the PENELOPE package, have been developed. The first code, penEasy, implements a modular, general-purpose main program and provides various source models and tallies that can be readily used to simulate a wide spectrum of problems. Its associated geometry routines, penVox, extend the standard PENELOPE geometry, based on quadric surfaces, to allow the definition of voxelised phantoms. This kind of phantoms can be generated using the information provided by a computed tomography and, therefore, penVox is convenient for simulating problems that require the use of the anatomy of real patients (e.g., radiotherapy treatment planning). The second code, penMesh, utilises closed triangle meshes to define the boundary of each simulated object. This approach, which is frequently used in computer graphics and computer-aided design, makes it possible to represent arbitrary surfaces and it is suitable for simulations requiring a high anatomical detail (e.g., medical imaging).
A set of software tools for the parallelisation of Monte Carlo simulations, clonEasy, has also been developed. These tools can reduce the simulation time by a factor that is roughly proportional to the number of processors available and, therefore,
