DIFFUSION SPECTRUM MRI OF STRUCTURAL CHANGES IN WHITE MATTER IN DEMENTIA

Northern California Institute/Res/Edu, San Francisco CA

Investigators

Linked publications & trials

Abstract

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. The overall goal of this project has been to develop robust methods for analysis of brain connectivity using diffusion spectrum imaging. We have made progress in adopting an approach that enables discovery of correlated structures with the least possible bias. The approach we devised is to analyze for any fiber pathway the set of all other pathways that meet it, i.e by sharing a voxel in common. This represents a simple definition of path-adjacency, equivalent to specifying a topology on the space of paths, and so is among the most basic of geometric properties. Path-adjacency includes familiar end-to-end connectivity as a special case, but also measures other aspects of path proximity in 3-dimensions, including path tangency and path crossing, and so it represents a highly unbiased probe of the 3D relational structure of the fiber pathways. We report the results of a new way of analyzing the fiber connectivity revealed by applying these methods to diffusion spectrum imaging (DSI) MRI in multiple primate species. Whereas previous tractography has suggested that fiber crossing is sporadic, adjacency reveals crossing to be to be pervasive and highly organized. We have discovered a mesh or grid of near right-angled fiber crossings as a basic unit of path organization. At each location, cerebral white matter is resolved into biaxial systems of intersecting fiber pathways, resembling the warp and weft of fabric. We term this an ortholinear structure. Thus, bi-axial structure at different depths is parallel - extrinsically parallel, inasmuch as the planes are locally parallel as 2D surfaces in 3D, and also intrinsically parallel, as each of the two fiber populations within each sheet is parallel to its counterpart in the other sheet. In many locations, a third set of pathways perpendicular to the other two is also observed. It appears the mesh network is continuous throughout the entire brain and not simply in discrete locations. These results reveal a new and simplified view of the pattern of connections that we have described as an ortholinear organization. It is an extension of the simple pattern of longitudinal, bilateral and dorso-ventral connectivity that characterizes the spinal cord and brainstem of all vertebrates. Further, known fiber pathways are all recognized within this ortholinear scheme, and adhere to it, but now are seen less as distinct structures than as local condensations of pathways of a particular orientation. This newly revealed organization represents an absolute, natural coordinate system of the brain. This natural coordinate system, like its instantiations in the more caudal segments of the CNS, likely expresses the logic and means of brain evolution and development, and will provide a natural framework description of brain structure and variation.