Disentangling Multispectral Functional Connectivity With Wavelets

Area of Science:

• Neuroscience
• Brain Imaging
• Network Science

Background:

• Brain connectomics aims to understand intrinsic brain organization through spontaneous activity.
• Functional connectivity (FC) is typically measured using linear correlations in BOLD-fMRI signals.
• BOLD-fMRI signals exhibit natural spectral scaling, suggesting multiscale analysis is beneficial.

Purpose of the Study:

• To apply wavelet transforms for analyzing multiscale functional connectivity in the brain.
• To investigate how brain networks develop across different spectral scales using BOLD-fMRI data.

Main Methods:

• Utilized wavelet transforms to analyze spontaneous BOLD-fMRI fluctuations.
• Employed information theoretic criteria to measure relatedness between spectrally-delimited FC graphs.
• Conducted voxelwise comparisons of graph structures across spectral bands.

Main Results:

• Demonstrated the utility of wavelet analysis for examining BOLD-fMRI connectivity at multiple scales.
• Identified the emergence of preferential functional networks across different spectral bands.
• Showcased how functional brain organization varies with spectral scale.

Conclusions:

• Wavelet analysis provides a powerful framework for exploring multiscale brain dynamics.
• Functional brain networks exhibit scale-dependent organization.
• This approach enhances understanding of the brain's intrinsic functional architecture.