Online Program Home
My Program

Abstract Details

Activity Number: 620 - Axles for Voxels: Recent Statistical Advances in Neuroimaging Data Analysis
Type: Topic Contributed
Date/Time: Thursday, August 2, 2018 : 8:30 AM to 10:20 AM
Sponsor: ENAR
Abstract #328365 Presentation
Title: Spatial Modeling of Diffusion Tensor Imaging Data from a Cocaine Addiction Study
Author(s): Dipankar Bandyopadhyay* and Zhou Lan and Brian Reich and Joseph Guinness
Companies: Virginia Commonwealth University and North Carolina State University and North Carolina State University and NC State University
Keywords: Bayesian; spatial process; Nearest Neighbor Gaussian process; Diffusion Tensor Imaging; Cocaine addiction; Markov chain Monte Carlo

In this talk, we propose a Bayesian spatial process model to capture spatial dependencies between proximal positive definite matrices (summary of DTI measures corresponding to each voxel), assuming the matrices to be marginally distributed as Wishart. An adaptation of the Nearest Neighbor Gaussian Process facilitates the corresponding Markov chain Monte Carlo computing framework. Simulation studies and applications to a real dataset on cocaine addiction illustrates the advantages of our method over available standard techniques in detecting brain hotspots.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program