Diffusion Spectrum MRI in Neural Fiber Structure: Quantitative Index Mapping and Applications

• Main Authors: Li-Wei Kuo, 郭立威
• Other Authors: Jyh-Horng Chen
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/74323274587696379808

碩士 === 國立臺灣大學 === 電機工程學研究所 === 91 === In neuroscience, it becomes an interesting topic to trace the fiber orientations in recent years. Diffusion spectrum imaging, DSI, is a new methodology to probe the 3-dimensional (3D) probability density function (PDF) of water molecular diffusion and map the neural fiber orientations by applying the 3D q-space theory. Differ from the diffusion tensor imaging, DTI, DSI can fully provide the information of the fiber-crossing and complex fiber structures. The accuracy of DSI in defining heterogeneous fiber orientations was validated and its ability to visualize complex cortical cytoarchitecture was demonstrated. In this study, we developed two quantitative indices, DSI mean squared length (MSL) and DSI anisotropy (DA), from DSI data. These indices extract the information of mean diffusive length and diffusion anisotropy of molecular diffusion. Both stroke and epilepsy of rat models are used to test these indices. Inherent contrast of these indices in different brain structures and the contrast-to-noise ratio (CNR) of the lesions were studied and compared with the DTI indices. Using DTI trace ADC and FA for comparison, this study tests the capability of DSI quantitative indices in differentiating normal structures and detecting lesions. Besides, we also discussed the difference between DSI and DTI in complex neural fiber structures. Non-Gaussianity of water molecular diffusion arising from intravoxel heterogeneity of fiber orientations may cause bias in DTI. We calculated the deviation angle between DSI primary vector and the DTI first eigenvector and compared them with the DSI complexity. From the results, we can get more understanding of these two non-invasive MR diffusion methods. In conclusion, we have derived DSI quantitative indices that extract the characteristics of PDF in a simple and quantitative way, providing a practical tool for neuroscience research and clinical application in the future. In the comparison of DSI and DTI, we can find that the accuracy of DTI was far from DSI in complex neural structures and may cause bias in orientation and index mapping. In the future, DSI will be the useful tool for neuroscience research and clinical applications. To understand the brain connectivity better.