Online Program Home
My Program

Abstract Details

Activity Number: 528 - Contributed Poster Presentations: Section on Statistics in Imaging
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Imaging
Abstract #306404
Title: A Local Group Differences Test for Subject-Level Multivariate Density Neuroimaging Outcomes
Author(s): Jordan Dworkin* and Kristin Linn and Theodore Satterthwaite and Armin Raznahan and Rohit Bakshi and Russell Shinohara
Companies: and University of Pennsylvania and University of Pennsylvania and Child Psychiatry Branch, National Institute of Mental Health, NIH and Harvard Medical School and University of Pennsylvania
Keywords: Neuroimaging; Multi-modal MRI; High-dimensional data
Abstract:

Much of neuroimaging research focuses on voxel-wise analysis, yet many diseases are characterized by diffuse processes that vary spatially across subjects. In simple cases these processes can be quantified using summary statistics of voxel intensities. Yet when the manifestation of a disease process is unknown or appears as a complex multi-modal relationship, summary statistics are often unable to capture group differences, and their use can encourage post-hoc searches for the optimal summary measure. Here we introduce a method for the naive discovery of group differences in voxel intensity profiles. The method operationalizes multi-modal MRI data as multivariate subject-level densities of voxel intensities, and develops a two-sample test for local points within the density space. We show that this method controls type I error and recovers relevant differences when applied to a single point, and demonstrate the ability to control family-wise error and maintain power when applying the test over a grid. Finally, we apply this method to a study of subjects with and without multiple sclerosis, and find significant differences in thalamic voxel intensity profiles.


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

Back to the full JSM 2019 program