2 resultados para Image Reconstruction

em ArchiMeD - Elektronische Publikationen der Universität Mainz - Alemanha


Relevância:

60.00% 60.00%

Publicador:

Resumo:

Die vorliegende Arbeit untersucht die Möglichkeiten und Limitierungen der MR-Lungenbildgebung mit hyperpolarisiertem 3-He bei gegenüber üblichen Magnetfeldstärken reduzierten magnetischen Führungsfeldern. Dabei werden insbesondere auch die funktionellen Bildgebungstechniken (dynamische Bildgebung, diffusionsgewichtete Bildgebung und Bestimmung des Sauerstoff-Partialdrucks) berücksichtigt. Experimentell geschieht dies durch in vivo Messungen an einem 0.1 T-Ganzkörpertomographen. Zur systematischen Untersuchung der MR-Bildgebung unter dem Einfluss diffundierender Kontrastmittel werden analytische Simulationen, Monte-Carlo-Studien und Referenzexperimente durchgeführt. Hier wird das Augenmerk besonders auf den Einfluss von Diffusions- und Suszeptibilitätsartefakten auf die morphologische und die diffusionsgewichtete Bildgebung gerichtet. Die Entwicklung und der Vergleich verschiedener Konzepte zur Erzeugung von MR-Führungsmagnetfeldern führt zur Erfindung eines neuartigen Prinzips zur weiträumigen Homogenisierung von Magnetfeldern. Die Umsetzung dieses Prinzips erfolgt in Form eines besonders kompakten Transportbehälters für kernspinpolarisierte Edelgase. Die Arbeit beinhaltet eine ausführliche Diskussion der MR-Bildgebungstechnik in Theorie und Praxis, um die Anknüpfungspunkte an die angestellten Untersuchungen herauszuarbeiten. Teile dieser Studien wurden von der Europäischen Raumfahrtorganisation ESA finanziert (Contract No.15308/01/NL/PA).

Relevância:

60.00% 60.00%

Publicador:

Resumo:

The subject of this thesis is in the area of Applied Mathematics known as Inverse Problems. Inverse problems are those where a set of measured data is analysed in order to get as much information as possible on a model which is assumed to represent a system in the real world. We study two inverse problems in the fields of classical and quantum physics: QCD condensates from tau-decay data and the inverse conductivity problem. Despite a concentrated effort by physicists extending over many years, an understanding of QCD from first principles continues to be elusive. Fortunately, data continues to appear which provide a rather direct probe of the inner workings of the strong interactions. We use a functional method which allows us to extract within rather general assumptions phenomenological parameters of QCD (the condensates) from a comparison of the time-like experimental data with asymptotic space-like results from theory. The price to be paid for the generality of assumptions is relatively large errors in the values of the extracted parameters. Although we do not claim that our method is superior to other approaches, we hope that our results lend additional confidence to the numerical results obtained with the help of methods based on QCD sum rules. EIT is a technology developed to image the electrical conductivity distribution of a conductive medium. The technique works by performing simultaneous measurements of direct or alternating electric currents and voltages on the boundary of an object. These are the data used by an image reconstruction algorithm to determine the electrical conductivity distribution within the object. In this thesis, two approaches of EIT image reconstruction are proposed. The first is based on reformulating the inverse problem in terms of integral equations. This method uses only a single set of measurements for the reconstruction. The second approach is an algorithm based on linearisation which uses more then one set of measurements. A promising result is that one can qualitatively reconstruct the conductivity inside the cross-section of a human chest. Even though the human volunteer is neither two-dimensional nor circular, such reconstructions can be useful in medical applications: monitoring for lung problems such as accumulating fluid or a collapsed lung and noninvasive monitoring of heart function and blood flow.