Meeting Banner
Abstract #2855

Composite Hypothesis Testing Via Support Vector Regression

Nagesh Adluru1, Richard J. Davidson1, Andrew L. Alexander1

1University of Wisconsin-Madison, Madison, WI, United States

Hypothesis testing is one of the most commonly performed statistical analyses in neuroimaging studies. Often the goal is to study the effects of variables such as physiological, behavioral and/or clinical on a variety of neuroimaging measurements e.g., cortical thickness, bold-activation, microstructural properties. A most common approach to test the significance of these effects is to project the measured neuroimaging data onto the linear models defined using the variables of interest via ordinary least squares. In this paper we propose that using a more robust projection can increase sensitivity to the effects and hence can also control for confounding parameters.

Keywords

accounts accurately adjusted advantages aims allow allows among analyses anxiety apply autism basic behavioral best better biologically brain called channel coil collinear composite context contiguous controls cook corrected covariance critical days defined demo detect diffusion disorders eddy either emotion encircled establish examine example field fractional freedom function gender general gnostic grating handedness head hormone hypothesis impact implementation implemented implicated important improvements leads learning least libs likelihood linear loss machine machines made maps mask matched matrix meaningful measures mens meters minim model models mostly nuisance null onto package packages paper parametric part physiological plays plots ponding popular popularly produces projected projections proposed pulse regression regulation reject related rely representation residuals resp robust role scanner serving setting shot significance since smith software sons spatially spin squared squares stat statistic statistical statistics stats steroid stress stud studies subjects superior support term terms threshold tool tract trio usually vantage vector wants weigh white whose year