|
Descripción
|
|
|---|---|
| Observed properties of functional brain networks strongly depend on what the network nodes depict. Therefore, the accurate definition of nodes is probably the most important methodological challenge of netowork neuroscience. In networks extracted from functional magnetic resonance imaging (fMRI) data, nodes are brain areas often referred to as Regions of Interest (ROIs); ROIs are collections of measurement voxels, typically defined a priori based on anatomical landmarks or functionality and connectivity observed in earlier studies. The ROI approach assumes functional homogeneity: all voxels of a ROI should have similar dynamics. Lack of functional homogeneity is interpreted as a technical flaw that should be fixed by introducing a novel set of ROIs. In my presentation, I will question this assumption. First, I will demonstrate that ROIs in three commonly-used fMRI parcellations show low levels of functional homogeneity. Therefore, they should not be used as nodes of functional brain networks without further consideration. Next, I will show that functional homogeneity of ROIs changes in time, reflecting changes in voxel-level connectivity inside ROIs. This will lead me to address the question about the functional meaning of these changes: I will demonstrate that functional homogeneity and internal connectivity structure of ROIs can predict their role in functional brain network topology. Inspired by these results, I will conclude by asking if any set of static ROIs can be optimal for constructing functional brain networks. | |
|
Internacional
|
No |
|
Entidad
|
Sciences Cognitives & Sciences Affectives @ Université de Lille, France |
|
Lugar
|
Centro de Tecnología Biomédica |
|
Páginas
|
|
|
Referencia/URL
|
|
|
Tipo de publicación
|
Ponencia |