3D simulation of external beam radiotherapy

Radiation therapy treatment is a very demanding cancer treatment process. The aim of the treatment is to cure or to limit the disease using high-energy radiation dose, having as minimum as possible damage on healthy tissues. In order to have the wanted results, the process is composed from several steps that are highly depended to each other. One could separate them into three different categories; the treatment planning and evaluation, the planning verification before and after treatment and finally the treatment itself. This work is a contribution in the chain of the radiotherapy process from the medical software application point of view. In principle the 3-Dimensional digital patient data are used instead of the physical patient, in order to perform the geometric planning (3D-Simulation), and partly the evaluation steps of the radiation therapy process. This thesis introduces the fully 3D definition of volumetric structures, offering higher degree of freedom to the clinicians for investigating the 3D information of the digital volumes. Also presents the semi-automatic segmentation of the spine, which is an essential tool used daily for the extraction of the spine volume. Main advantages of the methods are the increased efficiency and the improved time rates during volume segmentation. An interesting part of this work is the compensation of breathing artifacts as they are recorded and reconstructed on the surfaces of the acquired computed tomography medical volumes. This kind of artifacts is a source for potential errors during treatment planning and treatment evaluation. In this work we present a method for eliminating these inaccuracies and thus improving the treatment outcome. Further more using as basis the volume rendering pipe-line of the 3D-Simulation system reconstruction techniques have been established, to display anatomical volumes from specific body regions with sensitive structures. These methods require the minimum user interaction. Finally this thesis deals with a very essential issue related to the treatment planning verification. This is by presenting methods for the 3D visualization of the dose distribution in relation to the standard patient anatomy and the segmented anatomical structures.