Optimization of Magnetic Resonance Diffusion Tensor Imaging for Visualization and Quantification of Periprostatic Nerve Fibers
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Prostatectomy, surgical resection of the whole prostate is a common treatment for high- risk prostate cancer. Common side effects include long-time urinary and or erectile dysfunction due to damage inflicted to periprostatic nerves. The aim of this study was to identify an optimal magnetic resonance diffusion tensor imaging protocol for visualization and quantification of these nerves, as pre-surgery visualization may help nerve-sparing surgery. Both scanner filter, parameters for accelerated scan techniques, diffusion-related acquisition parameters and post- processing tractography parameters were investigated. Seven healthy volunteers were scanned with a state-of-art 3 T MRI scanner with varying protocol parameters. Diffusion data were processed and analysed using Matlab and Explore DTI. The resulting protocol recommendation included a normalized scanner filter, a parallel imaging acceleration factor of 2, partial Fourier sampling of 6/8, a right-left phase encoding direction, a b-value of 600 s/mm2, monopolar gradient polarity with applied eddy current correction, four acquisitions of 12 diffusion- sensitizing gradient directions, and a reverse phase encoding approach for correction of geometrical image distortions induced by static field inhomogeneity. For post-processing tractography, the recommended parameters were a lower limit for fractional anisotropy of 0.05, a minimum tract length of 3 centimetres and a maximum turning angle between voxels of 60 degrees. The limited parameter range that was tested and the low number of volunteers can be regarded as limitations to this study. Future work should address these issues. Furthermore, feasibility of periprostatic nerve tracking with the optimized protocol should be tested in a patient study.
Place, publisher, year, edition, pages
2015. , 79 p.
Diffusion tensor imaging, Diffusion-weighted imaging, Magnetic resonance imaging, Nerve-sparing prostatectomy, Prostate cancer, Quantitative analysis, Tractography
IdentifiersURN: urn:nbn:se:kth:diva-179658OAI: oai:DiVA.org:kth-179658DiVA: diva2:885493
Subject / course
Master of Science - Medical Engineering
Bjällmark, Anna, PhD
Nilsson, Mats, PhD