| As a specialized Magnetic Resonance Imaging (MRI) technique, diffusion imaging is sensitive to diffusion process of Nuclear Magnetic Resonance (NMR)-active nuclei in biological tissues. Studying the diffusion process of water molecules can reveal structural and environmental differences in normal versus diseased tissues at a microscopic (e.g. cellular) level. Diffusion-weighted magnetic resonance images (DWI) can be analyzed to obtain parameters such as Apparent Diffusion Coefficient (ADC), Fractional Anisotropy (FA), Radial Diffusivity (RD), Axial Diffusivity (AD), and others to be used as imaging markers in disease diagnosis and treatment assessment. To accurately obtain and reliably use these parameters, high spatial resolution with an adequate image quality must be achieved. Currently, the standard techniques for acquisition of diffusion images compromise on either image resolution or acquisition efficiency. The goal of this project is to develop a novel image acquisition technique for producing high-resolution and high-quality DWI within clinically acceptable scan duration (e.g. less than 3 minutes), and demonstrate its impact on the development of imaging markers for clinical applications. In order to accomplish this goal, three specific aims were defined.;In the first, we designed and implemented a new gradient and spin-echo (GRASE) based PROPELLER (periodically rotated overlapping parallel lines with enhanced reconstruction) pulse sequence, which we call "Steer-PROP". Experiments conducted for validating this sequence has proven that it is capable of producing high-resolution T2 and DWI in less than 2.5 minutes.;Second, we have identified the different kinds of phase errors that adversely impacted the image quality, and designed the corresponding correction methods to remove these phase errors during image reconstruction. Such phase correction has considerably improved the quality of diffusion-weighted images from the steer-PROP pulse sequence. Lastly, we have developed a software package, which we call Diffusion Imaging Visualization Environment (DIVE), for facilitating the calculation of a set of diffusion parameters that can be used to address clinical problems. The software provides a comprehensive diffusion image processing toolbox, capable of producing conventional parameters such as ADC and FA as well as a novel white matter characterization metric, which we call regional Fiber Coherence Index (r-FCI). The clinical significance of these parameters have been demonstrated through a diffusion tensor imaging study, designed to show the application of r-FCI and FA as markers for tumor cell infiltration along white matter fiber tracts. We have successfully completed this project and accomplished the goals defined in the three specific aims. We now have the necessary technical capabilities to produce high-resolution diffusion images that can be used to develop targeted imaging markers for specific clinical applications. |
