Skip to main content
. 2023 Oct 3;44(17):5729–5748. doi: 10.1002/hbm.26472

FIGURE 3.

FIGURE 3

Subject‐level estimation and the effect of smoothness. (a) Estimation of intrinsic connectivity network 6 (ICN 6) for a single subject from the Human Connectome Project (HCP) dataset using different data lengths and one subject from the Functional Imaging Biomedical Informatics Research Network (FBIRN). The subject‐level estimation is less smooth than the template obtained using group‐level analysis. The subject‐level estimate depends on several factors, including data characteristics such as length of data and original voxel size. (b) The smoothness of ICN 6 estimated as a function of data lengths for the HCP dataset is shown in green. The red color represents the same measure for the FBIRN dataset with 157 time points. Blue shows the smoothness level for the template of ICN 6. Y‐axis shows the normalized smoothness level with a maximum value of 1, which corresponds to a constant image (when all voxels have the same value, the normalized smoothness level is equal to one). Normalized smoothness equals one minus average gradient magnitude across the whole brain.