The influence of preprocessing steps on functional connectivity in resting state fMRI

摘要

Resting-state functional MRI (rs-fMRI) is an imaging technique providing the possibility to analyze the functional connectivity patterns of the human brain [1]. Functional connectivity can be evaluated through different methodologies. We can find for example Independent Component Analysis (ICA), Seed-based, Graph Theory (GT), etc. [2] [3]. In this study, we presented a several preprocessing strategies and we evaluated their influence on the functional connectivity. We have shown that the order of the preprocessing pipeline influence on spatial and temporal SNR as well as correlation matrix. The choice of the steps needed in preprocessing depends on the parametrical MRI acquisition, populations and methods used as post-processing. In this study we could recommend both strategies H and I, since they presented good compromising results.