Differences in functional connectivity distribution after transcranial direct-current stimulation: A connectivity...

Area of Science:

• Neuroimaging
• Statistical Analysis
• Biostatistics

Background:

• Functional connectivity (FC) analysis in neuroimaging often treats connectivity as individual edges.
• Relating complex FC patterns, viewed as statistical distributions, to clinical outcomes remains challenging.
• Existing methods may oversimplify the rich information within FC distributions.

Purpose of the Study:

• To develop and validate a novel statistical method for analyzing functional connectivity distributions.
• To apply this method to resting-state functional magnetic resonance imaging (rs-fMRI) data from a clinical trial.
• To investigate the impact of transcranial direct-current stimulation (tDCS) on brain connectivity in primary progressive aphasia (PPA).

Main Methods:

• Utilized estimated connectivity density between regions of interest as a functional covariate.
• Employed density quantiles instead of direct empirical density to enhance sensitivity to tail behavior.
• Applied a non-parametric, flexible approach to analyze rs-fMRI data from a tDCS intervention study in PPA patients.

Main Results:

• The analysis revealed that the tail of the connectivity density distribution, rather than mean or lower moments, significantly impacted classification outcomes.
• The novel method demonstrated effectiveness in identifying treatment effects in a PPA clinical trial.
• Significant differences in connectivity density tails were observed between tDCS and sham treatment arms post-intervention.

Conclusions:

• The proposed method offers a powerful, non-parametric approach to analyze functional connectivity as a distribution.
• Highlighting tail behavior of connectivity density improves the detection of subtle but significant effects.
• This approach reduces multiple comparisons and allows investigation of non-localized connectivity impacts, advancing neuroimaging analysis.