Affiliations: [a] Centre for Medical Image Computing, University College London, London, UK. [email protected] | [b] Department of Neuroscience, Brighton and Sussex Medical School, University of Sussex, Brighton, UK. [email protected]
Correspondence:
[*]
Address for correspondence: Department of Neuroscience, Brighton and Sussex Medical School, University of Sussex, Brighton, UK
Note: [†] Also affiliated at: Neuroimaging Laboratory, Santa Lucia Foundation, Rome.
Abstract: Mapping the brain structure and function is one of the hardest problems in science. Different image modalities, in particular the ones based on magnetic resonance imaging (MRI) can shed more light on how it is organised and how its functions unfold, but a theoretical framework is needed. In the last years, using network models and graph theory to represent the brain structure and function has become a major trend in neuroscience. In this review, we outline how network modelling has been used in neuroimaging, clarifying what are the underlying mathematical concepts and the consequent methodological choices. The major findings are then presented for structural, functional and multimodal applications. We conclude outlining what are still the current issues and the perspective for the immediate future.
