|
Abstract:
|
In neuroimaging, functional connectivity characterizes the dependencies between brain regions. Functional connectivity may be related to neurological disorders, gender, age, and other factors. However, it is challenging to compare images across populations due to high dimensionality. Joint and Individual Variation Explained (JIVE) seeks a lowrank approximation of the joint variation between two or more sets of features captured on common subjects and isolates this variation from that unique to each set of features. The analyses herein aim to evaluate the joint and individual variation between/within several datasets from the Philadelphia Neurodevelopmental Cohort study including neuroimaging, demographic measures, and psychosocial measures. We examine how joint and individual subject scores derived from neuroimaging relate to psychopathology, gender, and age. The analysis includes an examination of eigen-connectivity matrices that integrate full and partial correlation measures of functional connectivity. As several methods of implementing the JIVE framework have been developed recently, we compare their implementations in the aforementioned analyses.
|
