A new arXiv preprint introduces a generative model for brain connectivity. It focuses on structural connectivity (SC) and functional connectivity (FC), which provide complementary information about interactions between brain regions.
The proposed method is described as multimodal hypergraph-based flow matching. This is a generative modelling approach, though the abstract does not elaborate on the architecture or training details.
The authors note that SC and FC are widely used in neuroimaging studies of neuropsychiatric disorders. The potential value of the work is in generating these connectivity measures, but the abstract provides no results or comparisons.
With only one source, there are no conflicting findings to report. The contribution is a new way to generate both SC and FC together.