Learning the associations between brain connections and function/dysfunction
Stam Sotiropoulos1 & Theo Kypraios2
1 Sir Peter Mansfield Imaging Centre, School of Medicine, Division of Clinical Neuroscience
2 School of Mathematical Sciences
Summary
Understanding the workings of the human brain is one of the most outstanding challenges of our time. In particular, determining factors that contribute to the individual signature of integrated cognitive function is of genuine interest to neuroscience, but also of paramount importance for neurological applications; characterizing the “normal” brain structure and function is key for characterizing abnormalities and approaching disease mechanisms. Non-invasive and in-vivo magnetic resonance imaging (MRI), as well as Magneto-encephalography (MEG), can uniquely shed light to these questions.
This project will capitalise on advances and data offered by the cornerstone Human Connectome Project (HCP) (www.humanconnectome.org), for which the principal supervisor (SS) has been a major contributor. We will build novel computational methodology for estimating connections using complementary MRI and MEG through state-of-the-art inference techniques. In particular, we will develop models for estimating network structure from multimodal data and we will explore causal interactions. Using data-driven exploratory analysis, we will then identify latent associations between brain organisation and function. This will further allow the extraction of summary imaging-derived measures with certain contextual relevance that could comprise potential markers for subsequently exploring pathology-induced abnormalities and dysfunction. For instance, we will identify predictive behavioral traits of psychiatric disorders and explore their associations with estimated connectivity.
For funding details and applications please use:
http://www.nottingham.ac.uk/mathematics/prospective/research/maml.aspx