title: Tissue quantification based on Magnetic Resonance  Fingerprinting
creator: Rieger, Benedikt
subject: 004
subject: 004 Data processing Computer science
description: Quantification of tissue properties including the relaxation parameters has long been a goal of magnetic resonance imaging (MRI), to provide a basis for inter-patient comparability. However, extended acquisition times have hindered the usage of quantification for clinical applications. Magnetic Resonance Fingerprinting (MRF)was introduced as a promising method for simultaneous and fast quantification of multiple tissue parameters. Most MRF methods rely on spiral k-space trajectories, though they are well known to suffer from detrimental effects on the image quality, caused by gradient inaccuracies. The aim of thisworkwas to develop an implementation of the MRF paradigm for quantitative imaging based on Cartesian k-space readout, potentially increasing its usability and robustness. In a first step, a single slice MRF method based on echo-planar imaging (MRF-EPI) was developed, acquiring 160 gradient-spoiled EPI images with Cartesian readout.  By varying the flip angle, echo times and including an inversion pulse, fluctuating signal paths were created. T1 and T2* were quantified through matching the fingerprints with a precomputed dictionary. The quantification accuracy was validated in phantom scans showing good agreement  of MRF-EPI with reference measurements, with average deviations of 2+-3% and 2+-3% for T1 and T2*, respectively. In vivo maps were of high visual quality and comparable to in vivo reference measurements, despite the substantially shortened scan times of 10 s per slice. In a second step,  MRF-EPI was modified for improved volumetric coverage by using a slice-interleaved acquisition scheme. In addition to the T1 and T2* maps, proton density (PD) maps could be created without the need of additional measurements. In vivo whole-brain coverage of T1, T2* and PD with 32 slices were  acquired within 3:36 minutes, resulting in parameter maps of high visual quality and comparable performance with single-slice MRF-EPI at 4-fold scan-time reduction. In a final step the motion sensitivity of MRF methods was studied. Simulations demonstrated that MRF sequences based on spiral and Cartesian readout exert sensitivity to motion. To correct for motion, the individual measurements of MRF-EPI were corrected by co-registering them with an intensity-based coregistration method. Phantom and in vivo measurements demonstrated that motion artefacts were successfully mitigated with intensity-based co-registration, leading to motion-robust artefact-free T1 and T2* maps. Combining the developments of this work resulted in a fast and robust method for multi-parametric whole brain quantification in clinically acceptable scan time.
date: 2019
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/27275/1/Dissertation_Final.pdf
identifier: DOI:10.11588/heidok.00027275
identifier: urn:nbn:de:bsz:16-heidok-272758
identifier:   Rieger, Benedikt  (2019) Tissue quantification based on Magnetic Resonance Fingerprinting.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/27275/
rights: info:eu-repo/semantics/openAccess
rights: http://archiv.ub.uni-heidelberg.de/volltextserver/help/license_urhg.html
language: eng