An analytical workflow for seed-based correlation and independent component analysis in interventional resting-state...

Area of Science:

• Neuroscience
• Neuroimaging

Background:

• Resting-state functional MRI (rs-fMRI) identifies brain networks via correlated activity.
• Seed-based correlation analysis (SCA) and independent component analysis (ICA) are common but lack standardized workflows.
• Optimal analytical strategies for rs-fMRI are crucial for reliable results.

Purpose of the Study:

• To investigate and optimize analytical workflows for rodent rs-fMRI data using FSL.
• To compare ICA and SCA methods for resting-state network (RSN) analysis.
• To enhance the sensitivity and reliability of group comparisons in rs-fMRI.

Main Methods:

• Utilized rodent rs-fMRI data from longitudinal brain stimulation studies.
• Examined RSN identification and group comparisons in ICA.
• Compared ICA-based denoising with nuisance signal regression in SCA.
• Investigated seed selection strategies within SCA.

Main Results:

• In ICA, a baseline-only template improved functional connectivity and group difference detection compared to a pre/post stimulation template.
• In SCA, ICA-based denoising and individualised seeds increased sensitivity for detecting group differences.
• These methods prevented signal reduction, enhancing the detection of relevant group effects.

Conclusions:

• Employing baseline-only templates in ICA and ICA-based denoising with individualised seeds in SCA improves rs-fMRI analysis reliability.
• These optimized methods enhance result comparability across animal and human rs-fMRI studies.
• Standardized analytical workflows are essential for advancing rs-fMRI research.