January 13, 2010
Brain Imaging Mini workshops: David V. Smith and John A. Clithero (Duke U). 3pm Center for Advanced Brain Imaging Conference Room
John A. Clithero
Graduate Student, Huettel Lab
Department of Economics
Duke University
john.clithero@duke.edu
Workshop description:
Multivariate pattern analysis of neuroimaging data.
Workshop description:
Analyzing distributed patterns of brain activation using multivariate pattern analysis (MVPA) has become a popular approach to study functional magnetic resonance imaging (fMRI) data. This workshop will include both an overview of the methodology behind MVPA and an introduction to a new analysis package, PyMVPA (http://www.pymvpa.org/). PyMVPA, one of several freely available MVPA packages, is a Python module intended to ease pattern classification analyses of large datasets. Using recent examples from the literature, we will first outline what steps are required to suitably conduct MVPA on fMRI data, including feature selection, classifiers, sensitivity analysis, and data visualization. We will then explore how Python and PyMVPA can be used for each of these steps.
David V. Smith
Graduate Student, Huettel Lab
Department of Psychology & Neuroscience
Duke University
david.v.smith@duke.edu
Workshop title:
Using FSL for basic and advanced neuroimaging analyses
Workshop description:
Analysis techniques for functional magnetic resonance imaging (fMRI) have become increasingly complex over the past two decades, and several software packages have been developed to assist in analyzing these data. In this workshop, we will use FSL (http://www.fmrib.ox.ac.uk/fsl/), to analyze high-resolution fMRI data. We will first demonstrate how to preprocess data and implement B0 field maps to correct distortions arising from susceptibility artifacts in EPI data. Basic analyses will be performed using the general linear model (GLM); parametric extensions of these analyses will also be explored. Additionally, we will discuss advanced analysis techniques, including functional connectivity (e.g., PPI) and model-free analyses (e.g., ICA). Using model-free analysis methods, we will show examples of how this particular technique can be incorporated into preprocessing algorithms to remove spurious signals before statistical analyses.
Download David’s talk, documents and scripts
Download John’s talk Johns scripts are also loaded onto the Ubuntu Desktop in CABI 114.