title: Quantification of brain tissue oxygenation using magnetic resonance imaging - simulation and in vivo study
creator: Thomas, Sebastian Fritjof
subject: ddc-530
subject: 530 Physics
description: The oxygen consumption of the brain provides important information about its function and is of particular interest for therapy planning and monitoring in diseases such as brain tumors. Especially, the oxygen extraction fraction (OEF) is suitable as a biomarker because it is homogeneous in healthy tissue but develops contrast early in pathologies. The aim of this work was to develop and clinically apply a more robust magnetic resonance imaging (MRI)-based method for the determination of OEF in the brain. The method uses a mathematical model that depends not only on the OEF but also on other correlated parameters. Thus, the reconstruction is very sensitive to noise and is not applicable under clinical conditions. In this thesis, two new approaches were investigated: The first method  extends the normal fit process by a regularization term, which prefers a convergence of parameters in the physiologically meaningful range. The second method is based on an artificial neural network (ANN), which was trained on simulated data and subsequently used for the reconstruction of in vivo data. Both methods were able to produce more homogeneous maps in healthy subjects compared to the standard method. The regularization method found an OEF of (33.7 ± 2.0)% in gray matter (GM) and (34.2 ± 2.1)% in white matter (WM). The ANN method yielded an OEF of (35.2 ± 1.3)% in GM and (33.1 ± 1.7)% in WM. In two in vivo studies, the OEF was determined in smokers and in patients with brain tumors. In smokers, the OEF did not show significant differences both before and after smoking a cigarette and in comparison with never-smokers. The application in patients with brain tumors showed a significantly increased OEF in the tumor region compared to the contralateral side (p < 0.05). The presented methods led to more precise results under clinical conditions and may allow a clinical implementation of MRI-based oxygen measurement in the future.
date: 2020
type: Dissertation
type: info:eu-repo/semantics/doctoralThesis
type: NonPeerReviewed
format: application/pdf
identifier: https://archiv.ub.uni-heidelberg.de/volltextserverhttps://archiv.ub.uni-heidelberg.de/volltextserver/29144/1/Dissertation_Thomas_2020.pdf
identifier: DOI:10.11588/heidok.00029144
identifier: urn:nbn:de:bsz:16-heidok-291446
identifier:   Thomas, Sebastian Fritjof  (2020) Quantification of brain tissue oxygenation using magnetic resonance imaging - simulation and in vivo study.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/29144/
rights: info:eu-repo/semantics/openAccess
rights: http://archiv.ub.uni-heidelberg.de/volltextserver/help/license_urhg.html
language: eng