A Whole-Brain Computational Modeling Approach to Explain the Alterations in Resting-State Functional Connectivity during Progression of Alzheimers Disease

Abstract

Understanding the mechanisms behind Alzheimers disease (AD) is one of the most challenging problems in neuroscience. Recent efforts provided valuable insights on the genetic, biochemical and neuronal correlates of AD. The advances in structural and functional neuroimaging provided massive evidence for the AD related alterations in brain connectivity. In this study, we investigated the whole-brain resting state functional connectivity (FC) and variability in dynamic functional connectivity (v-FC) of the subjects with preclinical condition (PC), mild cognitive impairment (MCI) and Alzheimers disease (AD). The synchronization in the whole-brain was monotonously decreasing during the course of the progression. However, only in the AD group the reduced synchronization produced significant widespread effects in FC. Furthermore, we found elevated variability of FC in PC group, which was reversed in AD group. We proposed a whole-brain computational modeling approach to study the mechanisms behind these alterations. We estimated the effective connectivity (EC) between brain regions in the model to reproduce observed FC of each subject. First, we compared ECs between groups to identify the changes in underlying connectivity structure. We found that the significant EC changes were restricted to temporal lobe. Then, based on healthy control subjects we systematically manipulated the dynamics in the model to investigate its effect on FC. The model showed FC alterations similar to those observed in clinical groups providing a mechanistic explanation to AD progression.