My passion, and main focus in research, is in developing high spatial resolution models of the human brain structure. In a broad sense, these models fall in to two classes. One class characterizes the local structure of the human brain; that is, how any given point within the brain (1-2 mm cubic volume) is connected to its surrounding. The other class characterizes the global structure of the brain, that is, how distal points in the brain are connected to each other via long-range white matter fibers. Both these classes of models are represented in the form of a graph; a structure consisting of a series of vertices, and edges that connect the vertices to each other, which may be weighted to reflect the strength of connections.

These models can be readily studied to provide insight into local and global organisational principles of brain structure; to see how the brain structure of one individual varies from another, and how brain structure varies in normal ageing and in disease. But the added beauty is that these graph models can also be leveraged to provide insight in to brain function. This is done by interpreting acquired brain activity data as functions that reside on the vertices of the graphs. Brain activity data can then be processed and characterized in such way that is informed by the underlying structure on which it resides, in particular, using principles form the recently emerged field of graph signal processing.