Thesis (Ph. D. in Biomedical Engineering)--Vanderbilt University, Dec. 2004. / Title from title screen. Includes bibliographical references.

Proefschrift--Groningen. / Summary in Dutch and French.

Proefschrift--Groningen. / Summary in Dutch and French.

No description available.

No description available.

Functional proton-beam stereotactic radiosurgery requires sub-millimeter alignment accuracy. A patient tracking system called Sequential Alignment and Position Verification System (SAVPS) is under development at Loma Linda University Medical Center. An optical positioning system (OPS), manufactured by Vicon Peak, has been chosen to verify the correct alignment of the target with the proton beam axis. The main objective of this thesis is to optimize an existing version of SAVPS by conducting error analysis. An image processing algorithm was developed and applied to estimate the error introduced by the Patient Positioning System (PPS) in order to derive the true error of the SAVPS.

CAPES / Tratamentos de radiocirurgia requerem elevada precisão, pois envolvem lesões de pequenas dimensões, as quais são tratadas em poucas frações de altas doses de radiação. Sendo assim, a localização precisa da região de interesse é de grande importância para o sucesso do tratamento radioterápico. Neste trabalho foi investigada a interferência da caixa localizadora do alvo em tais procedimentos, utilizando imagens tomográficas do crânio de um phantom antropomórfico para simular um tratamento de radiocirurgia utilizando duas técnicas de exposição conforme protocolo do Hospital Israelita Albert Einstein. Todo o processo de planejamento radioterápico de um tratamento de radiocirurgia foi executado, incluindo o cálculo da dose média de radiação com e sem diferenças de heterogeneidades para todas as exposições realizadas, utilizando os algoritmos de cálculo de dose AAA e PBC. Foi realizada também uma análise da variância dos valores dos pixels nos histogramas dos números CT para analisar a interferência do uso da caixa de localização na qualidade das imagens tomográficas adquiridas. Os experimentos realizados indicam que o algoritmo AAA é menos susceptível a diferenças no cálculo dos valores das doses médias. Quanto ao ruído, os experimentos realizados com a caixa localizadora do alvo demostraram maior perda na qualidade da imagem tomográfica. / Radiosurgery treatments require high precision because they involve lesions of small dimensions, which are treated with elevated radiation doses in a few fractionated sessions. Thus, the localization accuracy of the region of interest is very important for successful radiation therapy. In this study, the interference of the target localizer box in such procedures was investigated using tomographic images of the skull of an anthropomorphic phantom in order to simulate a radiosurgery treatment using two exposure techniques according to a protocol of the Albert Einstein Hospital. The radiosurgery treatment process was performed, including the calculation of the average radiation dose with and without tissue inhomogeneity considerations for all exposures, using the AAA and PBC dose calculation algorithms. An analysis of the variance of pixel values in the CT number histogram was also conducted in order to analyze the interference of the target localization box on the quality of the acquired tomographic images. The experiments indicate that the AAA algorithm is less susceptible to differences in the calculation of the average doses. Regarding noise, the experiments performed with the target localization box demonstrated greater loss in tomographic image quality.