Diffusion tensor imaging (DTI) is a specialized MRI technique that is particularly suited to study muscle structure. Diffusion imaging provides quantitative information on muscle geometry, which is the main mechanical determinant of muscle performance, and on the local histopathological status of muscles to monitor disease progression or the effect of physical exercise or a pharmacological intervention. Diffusion MRI, like all other MRI techniques, is a noninvasive technology that provides a full three dimensional overview of the tissue of interest. It exploits the property that the apparent diffusivity of water is greatest along the dominant muscle fiber direction. In each imaging voxel a diffusion tensor is reconstructed from a series of diffusion-weighted MR images along at least 6 independent diffusion encoding directions. An eigenvalue analysis yields the principal axis of diffusion, which parallels the local muscle fiber orientation. Principal diffusion directions of neighboring voxels are combined for three-dimensional muscle fiber tractography. Most of the information gained by diffusion MRI is unique – in the sense that no alternative procedures exist that deliver similar insights. Diffusion tensor imaging (DTI) and fiber tractography of skeletal muscle is challenging for a number of reasons. First the short T2 relaxation times of muscles result in low SNR, which limits the resolution. Also the EPI readout leads to susceptibility-induced deformations in the images. Furthermore eddy currents and macroscopic motion can lead to a shift of the diffusion weighted volumes with respect to the non-weighted volume. The principal aim of the research described in the thesis was to optimize and standardize DTI data acquisition and processing and explore clinical application of the technique. For this purpose a custom built toolbox for Wolfram Mathematica 8 was developed, containing over 125 functions for simulating, processing, enhancing and analyzing DTI data. To investigate the clinical application of DTI a standardized protocol for acquiring and processing data was developed and tested to measure on various parts of the human body. Five studies were performed. The first study aimed to gain a better understanding of the influence of the sequence parameters, tensor-fitting algorithms, and post processing on image quality and SNR using a combination of simulations and experimentally acquired data. In the second study, reproducibility of the diffusion parameters was assessed. The third study focused on fiber tractography to visualize muscle architecture. In the fourth study DTI was applied to study delayed onset muscle soreness. In the final experimental study, the clinical feasibility of DTI in studies of the female pelvic floor was investigated.
|Qualification||Doctor of Philosophy|
|Award date||29 Oct 2012|
|Place of Publication||Eindhoven|
|Publication status||Published - 2012|