Statistics of brain functional networks: Classification and Testing

D. FRAIMAN; N. FRAIMAN; R. FRAIMAN

Congreso; SFN 2015; 2015

It is common today to describe neuroimaging data by using network or graph theory. Nevertheless, techniques for statistical analysis of these networks are not very developed. We herein address some statistical problems associated with the time evolution of brain functional networks constructed from EEG, MEG, or fMRI data. The approach presented here is non­parametric, i.e. it does not rely on a particular network model. Natural notions of center, variance and a depth function for networks that evolve in time are introduced. This allows us to develop several statistical techniques including testing, supervised and unsupervised classification, and a notion of principal component sets in the space of networks. Among other things, the results presented are important for the development of new diagnostic methods based on brain functional or structural network data. We show some examples, as well as a real data example.