False discovery rate controller for functional brain parcellation using resting-state fMRI

Parcellation of brain imaging data is desired for proper neurological interpretation in resting-state functional magnetic resonance imaging (rs-fMRI) data. Some methods require specifying a number of parcels and using model selection to determine the number of parcels with rs-fMRI data. However, this generalization does not fit with all subjects in a given dataset. A method has been proposed using parametric formulas for the distribution of modularity in random networks to determine the statistical significance between parcels. In this thesis, we propose an agglomerative clustering algorithm using parametric formulas for the distribution of modularity in random networks, coupled with a false discovery rate (FDR) controller to parcellate rsfMRI data. The proposed method controls the FDR to reduce the number of false positives and incorporates spatial information to ensure the regions are spatially contiguous. Simulations demonstrate that our proposed FDRcontrolled agglomerative clustering algorithm yields more accurate results when compared with existing methods. We applied our proposed method to a rs-fMRI dataset and found that it obtained higher reproducibility compared to the Ward hierarchical clustering method. Lastly, we compared the normalized total connectivity degree of each region within the motor network between normal subjects and Parkinson’s disease (PD) subjects using sub-regions defined by our proposed method and the entire region. We found that PD subjects without medication had a significant increase in functional connectivity compared to normal subjects in the right primary motor cortex using our sub-regions within the right primary motor cortex, whereas this significant increase was not found using the entire right primary motor cortex. These sub-regions are of great interest in studying the differences in functional connectivity between different neurological diseases, which can be used as biomarkers and may provide insight in severity of the disease.Applied Science, Faculty ofGraduat