Journal: bioRxiv

Article Title: Evidence for white matter intrinsic connectivity networks at rest and during a task: a large-scale study and templates

doi: 10.1101/2024.12.16.628798

Figure Lengend Snippet: Overview of our approach. The boxes numbered 1, 2, 5, and 6 indicate an open-source dataset (i.e. 900 ICs as well as 105 GM ICNs) and they helped us create a new robust and stable WM template with 97 ICNs (i.e., boxes 2, 6). The boxes 3 and 4 indicate the resting and task-based fMRI dataset we used to apply our template on using NeuroMark and GIFT scICA (indicated by box 7). Light blue boxes (8, 10, 15) represent the visualization of spatial maps to understand the summary of ICNs in WM and GM templates as well as after applying ICA to the BSNIP2 and MCIC datasets. We summarized the spatial maps using maximum intensity projections (MIPs) where we first normalized the dataset for each ICN and then computed the maximum along each voxel for the different ICNs, providing a single image which summarizes all the components. Note this was done separately for GM and WM ICNs. Purple boxes (9, 12, 14) indicate the processing that was done with the time courses for both BSNIP2 and MCIC datasets.

Article Snippet: These templates were derived using multi-model-order independent component analysis (ICA) ( , ), and then used as spatial priors within a spatially constrained ICA (scICA) pipeline called NeuroMark ( ; ).

Techniques: