Introduction

This website aims to be a tool to construct a functional network from fMRI signal and to do a graph theoretical analysis of the resulting network.

You will be able to upload your own fMRI data, of either healthy controls or patients, and quickly reconstruct the functional architecture associated with this data/network.

Each voxel in the fMRI corresponds to a node in the functional network and each links connects two voxels based on their functional interdependency (pairwise correlation).

Furthermore, this portal will perform several network analysis of the resulting functional brain architecture and will provide results on the importance of nodes (voxels) in the network. In this way, each node (voxel) will receive an associated score based on its importance according to different parameters, known in graph theory as “centralities”.

Data that you need to upload

All the files are expected to be in NIFTI format (.nii)

Structural Image: the MRI image of the brain

Functional Image: the functional MRI (fMRI) of the brain

Activation map: The resulting activation map image, after the fMRI statistical analysis has been performed (This is usually the image that radiologist look at, it is the fMRI with different brain area activated)

Correlation value: Correlation value threshold used for the fMRI statistical analysis, during the generation of the activation map (this is a tunable parameter adjusted by the radiologist while working on the activation map). Its value varies from 0 to 1. Leave it equal to 0 if unknown (0 value will make computation slower though).

Region of Interest (ROI): Minimum two ROIs are needed for the analysis. A ROI is a region of which you want to compute the functional connections (NIFTI file, .nii). Functional links are compute among all the voxels in the ROIs. Two voxels connects through a functional network if their interdependency is above a certain threshold.

Name: Each ROI should be named, so that results can be interpreted on the base of the names.

Calculate on your own functional network and perform network analysis.