pIMCo

pIMCo quantifies voxel-wise intermodal coupling across two or more imaging modalities using principal-component-based covariance decomposition to assess relationships among modalities such as cerebral blood flow, amplitude of low-frequency fluctuations, and local connectivity.

Key Features:

• Symmetric Coupling Coefficient: pIMCo defines a symmetric voxel-wise coupling coefficient via local covariance decompositions, contrasting earlier IMCo approaches that relied on asymmetric local weighted linear regression.
• Principal-Component-Based Decomposition: The method uses principal-component-based local covariance decomposition to summarize multivariate relationships among modalities at each voxel.
• Multi-Modality Compatibility: pIMCo generalizes intermodal coupling analysis to accommodate two or more imaging modalities.
• Voxel-wise Analysis: The approach computes coupling at the voxel level to capture fine-grained spatial patterns across modalities.

Scientific Applications:

• Neurodevelopmental Research: Applied to assess how intermodal coupling among cerebral blood flow, amplitude of low-frequency fluctuations, and local connectivity varies with age and sex in a cohort of 803 subjects aged 8-22.
• Disease Pattern Recognition: Used to identify multimodal coupling patterns that may serve as biomarkers or inform disease progression in neurological conditions.
• Large Cohort Multimodal Analysis: Enables analysis of complex multi-modal neuroimaging datasets to reveal patterns not evident within single modalities.

Methodology:

Uses principal-component-based local covariance decomposition to define a symmetric voxel-wise coupling coefficient applicable to two or more imaging modalities.