Activity Number:
|
342
- Novel Statistical Testing and Activation-Detection Methods for Imaging Data
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics in Imaging
|
Abstract #324977
|
|
Title:
|
Control of False-Positive Rates in Clusterwise fMRI Inferences
|
Author(s):
|
Jiangtao Gou* and Fengqing Zhang
|
Companies:
|
Hunter College of CUNY and Drexel University
|
Keywords:
|
fMRI ;
clusterwise inference ;
multiple test procedure ;
dependence structure
|
Abstract:
|
Eklund et al. (2016) question the validity of fMRI clusterwise inference, and they suggest that the main cause of the incorrect cluster inferences is the autocorrelation functions which do not follow the assumed squared exponential shape (Hayasaka and Nichols, 2003). Woo et al. (2014) also observe that large true positive clusters are likely to consist of mostly noise and cause the findings meaningless. We purpose solutions to correct the false-positive control in fMRI clusterwise inference by developing a semiparametric random field method to deal with the assumption violation of the squared exponential spatial autocorrelation function, and by developing a stepwise multiple test procedure with suitable dependence structures in neuroimaging data analysis.
|
Authors who are presenting talks have a * after their name.