A Multivariate Method for Estimating and comparing whole brain functional connectomes from fMRI and PET data

Area of Science:

• Neuroimaging
• Brain Connectomics
• Functional Neuroimaging

Background:

• Positron emission tomography (PET) and magnetic resonance imaging (MRI) are key for brain function visualization.
• Whole-brain functional connectomes from resting fMRI are common, but PET connectome analysis is less explored.
• Direct comparisons between whole-brain PET and fMRI connectomes are scarce.

Purpose of the Study:

• To develop and apply a spatially constrained independent component analysis (ICA) approach for estimating whole-brain PET connectomes.
• To compare PET-derived connectome patterns with those derived from resting-state fMRI.
• To explore the complementary information provided by PET and fMRI for biomarker development.

Main Methods:

• Utilized spatially constrained ICA to compute PET connectome estimates.
• Analyzed resting-state fMRI data to generate functional connectomes.
• Compared spatial patterns and network modularity between PET and fMRI connectomes.
• Examined PiB PET scans to identify comparable resting networks to fMRI.

Main Results:

• PET connectome analysis revealed highly modularized patterns complementary to fMRI connectomes.
• Comparable resting networks were identified from PiB PET scans, showing similar yet distinct spatial patterns to fMRI networks.
• PET networks exhibited slightly smoother spatial maps and occasional subnode variations compared to fMRI networks.

Conclusions:

• The developed PET connectome approach provides complementary insights to resting-state fMRI.
• Identified similarities and differences between PET and fMRI networks offer a new perspective on brain connectivity.
• This work paves the way for novel biomarker development by integrating information from both PET and fMRI modalities.