A Survey of Statistical Methods for Non-invasive Measurement of Connectivity in the Human Brain

Brian S. Caffo, Johns Hopkins University
Haley Hedlin, Johns Hopkins Department of Biostatistics
Suresh Joel, Johns Hopkins Kennedy Krieger Institute
Stewart Mofstovsky, Johns Hopkins Kennedy Krieger Institute
James Pekar, Johns Hopkins Kennedy Krieger Institute
Susan Spear-Bassett, Johns Hopkins Department of Psychiatry

Abstract

New developments in cognitive neuroscience focus on the human brain's functional integration. This is the study of the functional interactions, causes and the anatomical hierarchical circuitry of the brain, the variability in these quantities in the population and their impact on brain function and health. Functional connectivity in particular is defined as correlations between spatially remote neurological events. In this manuscript, we give a brief overview of singular value decomposition methods for statistically estimating brain networks associated with functional connectivity. We give a brief review of connectivity and eigen-value decompositions for estimating functional connectivity using functional magnetic resonance imaging.