Combined tracer distribution and background estimation for quasi-static tracer distributions in magnetic particle imaging
| Autor: | Straub, Marcel |
|---|---|
| Přispěvatelé: | Schulz, Volkmar, Stahl, Achim |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: | |
| Zdroj: | Aachen 1 Online-Ressource (xii, 132 Seiten) : Illustrationen, Diagramme (2018). doi:10.18154/RWTH-2018-227593 = Dissertation, RWTH Aachen University, 2018 |
| Popis: | Dissertation, RWTH Aachen University, 2018; Aachen 1 Online-Ressource (xii, 132 Seiten) : Illustrationen, Diagramme (2018). = Dissertation, RWTH Aachen University, 2018 Magnetic Particle Imaging (MPI) is a novel tomographic imaging modality that allows to reconstruct the spatial distribution of SuperparamagneticIron Oxide nanoparticles (SPIOs). It is a very fast, i.e. about 50 volumes per second, and highly sensitive imaging modality,i.e. few nanograms iron. By applying a magnetic excitation field, the magnetization of the SPIOs is continuously changed. The SPIOs can be detected by their non-linear magnetic response to the excitation field.A static magnetic gradient field, generating a Field Free Point (FFP),is superimposed. Only in the direct vicinity of the FFP, the excitation field is able to sufficiently change the magnetization of the SPIOs.Solely these SPIOs contribute to the measured signal. MPI uses three orthogonal excitation fields, which are strong enough to move the FFP through the required Field Of View (FOV). Hence, the excitation fieldsare often called drive fields. In this thesis the first experimental three-dimensional MPI device developed by Philips Research was rebuilt and recommissioned. During the recommissioning the power electronics as well as the receive path of the system have been thoroughly characterized and documented. Furthermore, a full software stack for data processing, image reconstruction and system simulation was developed. After the successful recommissioning, a slowly changing background component of the measured data was identified as a challenge for the post-processing and reconstruction. Hence, a method was developed to reduce the effect of the changing background without interrupting the measurement of a sample. To this purpose, the data acquisition was modified to apply a small spatial shift of the FOV between consecutively recorded volumes. The background is assumed to be static, whereas the sample is shifted. A reconstruction algorithm based on a constrained least squares fit was developed to separate the image and the background from the shifted measurements. The modified acquisition and reconstruction scheme was successfully evaluated by simulations as well as measurements. The Signal to Background Ratio (SBR) of the reconstructed images was improved by a factor of ten (10) in simulations and a factor of two (2) in actual measurements. Published by Aachen |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|