Quantitative FMRI Based on Intrinsic Connectivity Network Atlasing: Application to Longitudinal Data in Healthy Controls

Abstract

Understanding the brain's complex connectivity patterns during cognitive tasks still a significant challenge in neuroscience despite significant advancements in functional magnetic resonance imaging (fMRI) technology. This thesis investigates the stability and engagement of Intrinsic Connectivity Networks (ICNs) during face and word encoding tasks in healthy individuals, with an aim to evaluate the reproducibility of these patterns over time. While much of Traditional fMRI analysis has has focused on task-based paradigms and specific brain regions, potentially overlooking brain activity's continuous and dynamic nature. The study uses the ICN_Atlas, an innovative tool to quantify intrinsic connectivity patterns and analyze longitudinal fMRI data across multiple time points. Findings show that certain ICNs, especially those associated with visual and motor/visuospatial functions, exhibit consistent engagement during face and word encoding tasks. This consistency was observed across baseline, 3-month, and 1-year follow-ups, which suggest that the neural mechanisms involved in these tasks are stable over time. Despite the study's limited number of participants, the results offer valuable insights into the reproducibility of fMRI based ICN engagement metrics. The research underscores the importance of considering intrinsic connectivity patterns in cognitive neuroscience and suggests that the ICN_Atlas could be a game-changer in future studies with respect to understanding brain function. The thesis also highlights that more study is required to track these patterns in clinical populations, particularly in relation to neurological disorder.