American Statistical Association
Functional Magnetic Resonance Imaging (fMRI) data are used routinely by practitioners in neuroimaging to obtain networks in the brain that correspond to different brain functions. Recently, a number of important initiatives have emerged where data from various centers worldwide have been gathered in publicly available repositories in an effort to enhance exploratory analysis of imaging data. The resulting datasets are massive and heterogeneous. In this talk, I will describe a matrix decomposition method for the analysis of large datasets for obtaining subject-specific mixing matrices and common brain networks. In addition, issues with the preprocessing of imaging data that can affect the statistical analyses of the data will be discussed.
|Date:||Tuesday, October 21, 2014|
“Central Park” meeting area at the Child Study Center
One Park Avenue (between 32nd and 33rd Streets)
New York, New York