Online Program Home
My Program

Abstract Details

Activity Number: 659
Type: Contributed
Date/Time: Thursday, August 4, 2016 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics in Imaging
Abstract #319768 View Presentation
Title: Including Data-Analytical Stability in Cluster-Based Inference
Author(s): Sanne Roels* and Tom Loeys and Beatrijs Moerkerke
Companies: Ghent University and Ghent University and Ghent University
Keywords: fMRI ; reproducibility ; cluster inference

A big challenge in the statistical analysis of functional Magnetic Resonance Imaging (fMRI) data is to account for simultaneously testing activation in over 100.000 volume units or voxels. A popular method that reduces the dimensionality of this test problem is cluster-based inference. We propose a new testing procedure that allows to control the family-wise error (FWE) rate at cluster level but improves cluster-based test decisions in two ways by (1) taking into account a measure for data-analytical stability and (2) allowing voxel-based interpretation of results. For each voxel, we define the re-selection rate conditional on a given threshold and add this as a measure for stability into the selection process. Our procedure distinguishes between a soft and hard FWE controlling threshold. Clusters that survive the soft but not the hard criterion get selected if sufficient evidence for voxelwise stability is present. Using the Human Connectome Project Data, we demonstrate how in a group analysis our method results in a higher number of selected voxels but also in larger overlap between different test images. We found that the procedure is more advantageous for small sample sizes.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association