Resting State Network Detection with Searchlight on Functional MRI
Shiyang Chen 1,2 , Hasan Ertan Cetingul 2 , Xiaoping Hu 1,3 , and Mariappan S. Nadar 2
The Wallace H. Coulter Department of
Biomedical Engineering, Georgia Institute of Technology
and Emory University, Atlanta, GA, United States,
and Computer Vision, Siemens Corporation, Corporate
Technology, Princeton, NJ, United States,
Imaging Technology Center, Emory University, Atlanta,
GA, United States
Resting state networks (RSNs) are important biomarkers
for disease diagnosis, (e.g. Alzheimers disease). The
conventional methods to detect RSNs require a spatial
smoothing step to compensate the low SNR of fMRI, and
they assume the brain connectivity is voxel-to-voxel.
These steps and assumptions are still controversial. We
propose a searchlight plus multivariate regression
method to detect the RSNs, which is able to detect
region-to-region brain connection and can be performed
on unsmoothed or slightly smoothed fMRI data. We
validate this method by detecting the conventional RSNs,
and we also found additional brain regions that are
connected in the RSNs.
This abstract and the presentation materials are available to members only;
a login is required.