DocumentCode
3494719
Title
Peak Geodesic Concentration: A measure of WM complexity in HARDI
Author
Rao, Arvind ; Smith, Alex ; Schultz, Robert ; Roberts, Timothy ; Verma, Ragini
Author_Institution
Dept. of Radiol., Univ. of Pennsylvania, Philadelphia, PA, USA
fYear
2012
fDate
9-10 Jan. 2012
Firstpage
233
Lastpage
238
Abstract
We design and evaluate a scalar measure for HARDI data called Geodesic Concentration (GC) that characterizes the integrity of white matter (WM), by quantifying the anisotropic concentration of water diffusion. Mathematically, GC is defined as the concentration relative to the peak, identified in the HARDI model fitted to the data. As GC reflects the degree of directionally coherent diffusion, it is sensitive to the presence of single or multiple fiber populations within a voxel and hence is able to characterize complexity of WM, yielding a unique HARDI based scalar measure that can be used for evaluation of pathology. GC properties are elucidated with the help of simulated data and its applicability to population studies is evaluated on a population of autism patients and controls.
Keywords
biodiffusion; biomedical MRI; brain; differential geometry; diseases; medical image processing; neurophysiology; physiological models; HARDI data; HARDI model; WM complexity; anisotropic concentration; autism patients; directionally coherent diffusion; high angular resolution diffusion imaging; multiple fiber populations; pathology; peak geodesic concentration; scalar measurement; single fiber populations; unique HARDI based scalar measurement; water diffusion; white matter; Complexity theory; Equations; Harmonic analysis; Level measurement; Mathematical model; Optical fiber dispersion; Variable speed drives;
fLanguage
English
Publisher
ieee
Conference_Titel
Mathematical Methods in Biomedical Image Analysis (MMBIA), 2012 IEEE Workshop on
Conference_Location
Breckenridge, CO
Print_ISBN
978-1-4673-0352-1
Electronic_ISBN
978-1-4673-0353-8
Type
conf
DOI
10.1109/MMBIA.2012.6164756
Filename
6164756
Link To Document