Home Research Teaching Personal

  Main   Mathematical Models in Computer Vision: The Handbook, Springer (2005)

  • Characterization of Diffusion Anisotropy in DWI, Y.Chen
ABSTRACT: Diffusion-weighted magnetic resonance imaging (DWI) is unrivaled in its ability to quantify changes in biological tissue microstructure noninvasively. The quantification is based on the  anisotropy of water diffusion  and fiber pathways determined from DWI measurements. This chapter is devoted to the study of the characterization of diffusion anisotropy. Two methods for characterizing  diffusion anisotropy are introduced. One uses transition  probability density function (PDF), and the other uses  apparent diffusion coefficient (ADC) profiles. Techniques for estimating  the PDF and  ADC profiles from high angular resolution DWI are reviewed. In particular we  presented a  variational framework  for the  estimation of the PDF modeled as a mixture of two Gaussians.  We also described a variational model for the estimation of the ADC profiles represented by a truncated spherical harmonic series, and the algorithm for the  characterization of diffusion anisotropy using  ADC profiles. These two models are distinguished by simultaneous smoothing and estimation. Experimental results indicate the effectiveness of these models in enhancing and revealing intravoxel information.

Last Update: December20th, 2004, you can mail your comments to: nikos.paragios@computer.org