Abstract
We introduce a connectomics metric that integrates information on structural connectivity (SC) from diffusion MRI tractography and functional connectivity (FC) from resting-state functional MRI, at individual subject level. The metric is based on the ability of SC to broadly predict FC using a simple linear predictive model; for each connection in the brain, the metric quantifies the deviation from that model. For the metric to capture underlying physiological properties, we minimise systematic measurement errors and processing biases in both SC and FC, and address several challenges with the joint analysis. This also includes a data-driven normalisation approach. The combined metric may provide new information by indirectly assessing white matter structural properties that cannot be inferred from diffusion MRI alone, and/or complex interregional neural interactions that cannot be inferred from functional MRI alone. To demonstrate the utility of the metric, we used young adult data from the Human Connectome Project to examine all bilateral pairs of ipsilateral connections, i.e. each left-hemisphere connection in the brain was paired with its right-hemisphere homologue. We detected a minority of bilateral pairs where the metric value is significantly different across hemispheres, which we suggest reflects cases of ipsilateral connections that have distinct functional specialisation in each hemisphere. The pairs with significant effects spanned all cortical lobes, and also included several cortico-subcortical connections. Our findings highlight the potential in a joint analysis of structural and functional measures of connectivity, both for clinical applications and to help in the interpretation of results from standard functional connectivity analysis.
Significance Statement Based on the notion that structure predicts function, the scientific community sought to demonstrate that structural information on fibre bundles that connect brain regions is sufficient to estimate the strength of interregional interactions. However, an accurate prediction using MRI has proved elusive. This paper posits that the failure to predict function from structure originates from limitations in measurement or interpretation of either diffusion MRI (to assess fibre bundles), fMRI (to assess functional interactions), or both. We show that these limitations can be nevertheless beneficial, as the extent of divergence between the two modalities may reflect hard-to-measure properties of interregional connections, such as their functional role in the brain. This provides many insights, including into the division of labour between hemispheres.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
↵* Oren Civier. Email: orenciv{at}gmail.com
