Tract-based probability densities of diffusivity measures in DT-MRI

Authors:
Çağatay Demiralp;David H. Laidlaw
Affiliations:
Brown University;Brown University
Venue:
MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part I
Year:
2010

Citing 7
Cited 0

Elements of information theory

Elements of information theory
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
Permutation, Parametric, and Bootstrap Tests of Hypotheses (Springer Series in Statistics)

Permutation, Parametric, and Bootstrap Tests of Hypotheses (Springer Series in Statistics)
Group Statistics of DTI Fiber Bundles Using Spatial Functions of Tensor Measures

MICCAI '08 Proceedings of the 11th international conference on Medical Image Computing and Computer-Assisted Intervention - Part I
Multivariate High-Dimensional Cortical Folding Analysis, Combining Complexity and Shape, in Neonates with Congenital Heart Disease

IPMI '09 Proceedings of the 21st International Conference on Information Processing in Medical Imaging
A Tract-Specific Framework for White Matter Morphometry Combining Macroscopic and Microscopic Tract Features

MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part II
Tract-based morphometry

MICCAI'07 Proceedings of the 10th international conference on Medical image computing and computer-assisted intervention

Quantified Score

Hi-index	0.00

Visualization

Abstract

We evaluate probability density functions of diffusivity measures in DTI fiber tracts as biomarkers. For this, we estimate univariate and bivariate densities, such as joint probability densities of the tract arc length and FA, MD, RD, and AD, in the transcallosal fibers in the brain. We demonstrate the utility of estimated densities in hypothesis testing of differences between a group of patients with VCI and a control group. We also use the estimated densities in classifying individual subjects in these two groups. Results show that these estimates and derived quantities, such as entropy, can detect group differences with high statistical power as well as help obtain low classification errors.