Investigating Brain Function using Functional MRI-based Meta-analysis and Diffusion Tensor Imaging

Abstract

The investigation of structure and function in biological systems, specifically the brain, is paramount given the close relationship that they share with each other. Functional magnetic resonance imaging (fMRI) and Diffusion Tensor Imaging (DTI) are two non-invasive MR-based modalities which are popular for investigating function and structure, respectively. In this thesis, we present novel advances in fMRI and DTI post-processing which will likely hasten the widespread use of these modalities for understanding brain function. Specifically, in the first application, Activation Likelihood Estimation meta-analyses (a popular statistical method used for assimilating results across many fMRI studies) along with meta-analytic connectivity modeling and DTI were assembled in a novel analysis pipeline in order to identify the neural substrates underlying gender differences in all forms of suicidal behavior in an attempt to form specific new hypotheses which can be tested in future experimental studies. In the second application, DTI was used in a dog model for generating a voxel-specific atlas of tensor (and hence, axonal) orientations by adapting existing human pipelines to dog data. Further, probabilistic tractography was utilized to test the hypothesis that previously seen anterior-posterior dissociation in the Default Mode Network (DMN) in dogs in a resting state fMRI study could have a structural basis. Contrasting anterior-posterior DMN connectivity in dogs with that in humans may help us understand the evolutionary role of the DMN.