MRI - Nuclear magnetic resonance imaging has since its beginnings, to present time developed significantly, which can be seen also in new imaging techniques and their applications in clinical diagnostics. Static magnetic fields are harmless to human body and MRI also does not produce any ionizing radiation. Therefore, MRI is considered harmless to humans.
One of important and frequently used MR imaging methods is imaging of anisotropic diffusion. Many tissues show anisotropic diffusion of water molecules. Typical example of anisotropic tissue are nerve fibers, where diffusion is faster in direction parallel to fibers than in direction perpendicular to fibers. Measuring diffusion constants in different directions enables us to find the direction of fastest diffusion and therefore fiber direction. If we measure diffusion in at least six different directions in space, we can use the data to calculate the diffusion tensor.
In this thesis, we first study the effect of diffusion on the MRI signal and the basics of measuring Apparent Diffusion Coefficients (ADC) with the imaging method of Pulsed Gradient Spin Echo (PGSE). Firstly, we implement PGSE on a system for high spatial resolution MR imaging at Jožef Stefan Institute and measure ADC of different plant samples and of water. Secondly, we upgrade the implemented method to Diffusion Tensor Imaging (DTI) in 3D and use the method to measure DTI of mouse brain. Finally, we implement an algorithm for diffusion tensor calculation and show orientation of nerve fibers in mouse brain. In the thesis we also analyze how errors of diffusion tensor calculation depend on the selection of the number and direction of different diffusion gradients. We also analyze influence of induced internal magnetic field gradients on the precision of ADC measurments.