Some Likelihood Ratio Tests for Signal Detection from fMRI Brain Images

• Activity Number: 249
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Khalil Shafie*
• Companies: University of Northern Colorado
• Keywords: likelihood ratio statistic ; scale space random fields ; rotation space random fields ; equivalent Gaussian measures ; reproducing kernel Hilbert space

Abstract

For the past two decades, Gaussian random field theory has been extensively used to model the brain images. In this work, we utilize two approaches to drive the likelihood ratio test statistic for some different models used for activation signal detection in fMRI (functional magnetic resonance imaging). In the first approach we use the reproducing kernel Hilbert space (RKHS) machinery and in the second one we employ the techniques of Gaussian measures on Hilbert spaces. The models considered have the form of smoothed version of signal plus a white noise which include scale and rotation space random fields with one or more signals as special cases.