Have you seen the Kaggle competition on brain functional MRI (fMRI)? It was called: MLSP 2014 Schizophrenia Classification Challenge. The number of public databases of neuroimaging data is increasing and with them, the number of scientific open questions regarding its pre- and post-processing.
They offered 2 types of features: Functional Network Connectivity (FNC) and Source-Based Morphometry (SBM) loadings. These features are obtained through a pre-processing pipeline that can be rather complex and still presents many open scientific questions.
I will show in this talk some tools on how to deal with this type of data, how to use Python to do most of the pre-processing of one subject’s fMRI data, and one way to create a functional connectivity matrix.
You can also find the slides here