Meeting Banner
Abstract #3664

Using Atrophy as a Marker of Disease Severity to Understand the Evolution of DTI Changes in Alzheimer&[prime]s Disease

Julio Acosta-Cabronero1, Dina Kronhaus2, Robert J. Arnold1, Guy B. Williams3, Peter J. Nestor4

1Department of Clinical Neurosciences, University of Cambridge, Cambridge, Cambridgeshire, United Kingdom; 2Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, Cambridgeshire, United Kingdom; 3Wolfson Brain Imaging Centre, University of Cambridge, Cambridge, Cambridgeshire, United Kingdom; 4Plasticity and Neurodegeneration, German Center for Neurodegenerative Diseases (DZNE), Magdeburg, Saxony-Anhalt, Germany

This study explored the hypothesis that diffusion tensor imaging (DTI) abnormalities in Alzheimers disease (AD) should capture axonal damage or cell death. This was done by testing the linear dependence of DTI-derived metrics on hippocampal atrophy using whole-brain and region of interest approaches. The results revealed that in the same white matter areas where axial diffusivity (&[lambda]1) is abnormal early, progressive change in radial diffusivity (RD) and fractional anisotropy (FA) but not in &[lambda]1 predict atrophy in the hippocampus. What is then &[lambda]1 sensitive to in very early disease stages?&[prime]

Abstract PDF Presentation Video

Keywords

according additional albeit alterations anatomical application applied atrophy axial axis behaves better bilaterally boundaries brain capture capturing caudal clinic clinical coefficients cognition cognitive coil collinear computer connects coronal corpus correct correlated correlations criteria deficits definition degeneration dementia determined diffusion disease early eddy equal event evidence evolution explored extensive forty free global gradients head identify implying incipient inconsistent independent independently jack kingdom length linear lobe loss manually maps marked marker matched mathematics matrix measure measured mechanisms memory metrics models network neural neuronal never note orientations overall parietal patients permutations peter physics plasticity posterior previously processes pulse question recruited reduced reductionist reductions reflect refocused regional regressions related relationships resolution revealed session several severity shot shrinkage significance simulations slices slightly sometimes stages statistical strongest strongly subdivisions subject suggests superior supporting survive system temporal tensor theoretical threshold thresholded toolbox traced tracts trio twice unbiased understand upstream validated vary vitro volume volumes whatever whereas white whole