Diffusion Tensor MRI Analysis for Early Detection of AD

~-- -------------- Diffusion Tensor MRI Analysis for Early Detection of AD Jun Zhang ABSTRACT This project will develop robust diffusion tensor imaging analysis techniques to provide computer-based automatic tools and means for early detection of Alzheimer's disease (AD). While anatomical changes are evident in the brains of AD patients late in the disease the earliest indications of the disease have come from diffusion tensor measurements. Yet these results are complicated by the inability of traditional diffusion tensor magnetic resonance imaging (DT-MRI) and analysis methods to resolve changes in the diffusion characteristics of the white matter where individual voxels contain white matter fiber tracks with multiple directions. We have developed an alternative analysis method based on a direct solution of the diffusion equation which we believe provides an unambiguous resolution of the problem of resolving tracks with crossing fibers. The hypothesis of this work is that our new tracking method will enable a more sensitive and therefore earlier detection of the pathological changes accompanying the onset of AD. The overall goal of this research work is to develop diffusion tensor imaging and computational analysis algorithms to differentiate crucial white matter fiber tracking in the temporal lobes of brains - using DT-MRI. This goal is much broader than previous DT-MRI-related works on AD carried by others, in which some diffusion-based parameters in certain regions of interest are measured and are compared. In particular, the developed algorithms will be used to reconstruct white matter fiber tracts from the cingulate gyrus in the brains of both early AD patients and in the age-matched healthy controls. The 34 DT-MRI datasets that will be used to validate the developed algorithms are provided by the PI's collaborator from Australia. Some very encouraging preliminary results have been obtained by using the current algorithm and analysis tools on both AD and healthy control datasets. Dr. Jun Zhang and his research team and colleagues will work collaboratively to develop robust and reliable fiber tracking algorithms based on the diffusion equation approach to reconstruct certain white matter fiber tracts that are known to be affected by AD in its early stage. The developed tools will include AD-tailored fiber tracking algorithm, software for colored 3D tract visualization, and methods for statistical data analysis. These tools will be used to search for a DT-MRI-based noninvasive in vivo biomarker in the context of monitoring the change of certain white matter fiber tracts for early detection of AD. An accurate and reliable noninvasive means of detecting early AD or predicting AD will help scientists and doctors develop very early treatments to prevent or delay the onset of AD.