Ranking fMRI time courses by minimum spanning trees: assessing coactivation in fMRI

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Data Analysis

Background:

• Functional magnetic resonance imaging (fMRI) reveals brain activity through time courses.
• Similar temporal activation patterns in fMRI time courses can indicate functional connectivity or coactivation between brain regions.
• Assessing the homogeneity of these time course groups is crucial for accurate interpretation.

Purpose of the Study:

• To introduce and demonstrate the utility of minimum spanning tree (MST) ranking for analyzing temporal homogeneity in fMRI time courses.
• To show how MST ranking can be applied to both data-driven and hypothesis-led fMRI analysis methods.
• To enable pairwise comparisons between different groups or clusters of fMRI time courses.

Main Methods:

• Utilizing minimum spanning tree (MST) to order multidimensional fMRI time courses.
• Applying MST ranking to assess the self-consistency (homogeneity) of temporal activation patterns within groups of time courses.
• Developing a method for pairwise comparisons of fMRI time course groups.

Main Results:

• MST ranking effectively investigates the temporal homogeneity of fMRI time course groups.
• The method is applicable to data-driven approaches for identifying coactivation in fMRI.
• MST ranking is also valuable for hypothesis-led analyses and facilitates group comparisons.

Conclusions:

• MST ranking provides a robust framework for analyzing temporal dynamics and functional connectivity in fMRI data.
• This approach enhances the investigation of coactivation patterns and allows for flexible group comparisons.
• The method offers a novel way to explore the relationships within and between sets of fMRI time courses.