Functional connectivity analysis of fMRI data using parameterized regions-of-interest

Area of Science:

• Neuroimaging
• Functional Magnetic Resonance Imaging (fMRI)
• Computational Neuroscience

Background:

• Accurate functional connectivity analysis in fMRI relies on precise region-of-interest (ROI) definition.
• Traditional ROI selection methods, such as those based on General Linear Models (GLM) or anatomical atlases, face limitations including conservative multiple comparison corrections and biased structure-function mapping.
• These limitations can hinder the accurate identification and analysis of functional brain networks.

Purpose of the Study:

• To propose a novel method for defining functional ROIs in fMRI data that bypasses the need for stringent multiple comparison correction.
• To extend the Activated Region Fitting (ARF) framework for robust functional connectivity estimation.
• To demonstrate the efficacy of the proposed method in recovering brain region connectivity through simulations and real-world data application.

Main Methods:

• The study extends the Activated Region Fitting (ARF) framework, originally developed for fMRI analysis, to facilitate connectivity analysis.
• ARF models entire fMRI data volumes using a limited set of parameters, defining regions of activation.
• This parameter-driven approach necessitates a less stringent multiple comparison procedure, allowing direct use of activation regions for functional connectivity estimation.

Main Results:

• Simulations demonstrated that the Activated Region Fitting method can successfully recover the connectivity patterns between brain regions.
• The application of the method to real fMRI data from a Go/No-Go experiment showcased its practical advantages.
• The proposed approach effectively defines functional ROIs, enabling more reliable connectivity analyses.

Conclusions:

• Activated Region Fitting provides an effective solution for defining functional ROIs in fMRI data, overcoming limitations of conventional methods.
• The method simplifies the process of functional connectivity analysis by reducing the stringency of multiple comparison corrections.
• This approach offers a valuable tool for neuroimaging research, enhancing the accuracy and efficiency of brain connectivity studies.