Connectogram

With the recent concerted push to map all of the human brain and its connections,[8][9] it has become increasingly important to find ways to graphically represent the large amounts of data involved in connectomics. Most other representations of the connectome use 3 dimensions, and therefore require an interactive graphical user interface.[1] The connectogram can display 83 cortical regions within each hemisphere, and visually display which areas are structurally connected, all on a flat surface. It is therefore conveniently filed in patient records, or to display in print. The graphs were originally developed using the visualization tool called Circos,.[10][11]

Connectogram, typical of those in clinical use, depicting estimated connection damage in Phineas Gage, who in 1848 survived a large iron bar being propelled through his skull and brain. The connectogram shows only the connections that were estimated to be damaged.

On an individual level, connectograms can be used to inform the treatment of patients with neuroanatomical abnormalities. Connectograms have been used to monitor the progression of neurological recovery of patients who suffered a traumatic brain injury (TBI).[12] They have also been applied to famous patient Phineas Gage, to estimate damage to his neural network (as well as the damage at the cortical level—the primary focus of earlier studies on Gage).[13]

Connectograms can represent the averages of cortical metrics (grey matter volume, surface area, cortical thickness, curvature, and degree of connectivity), as well as tractography data, such as the average densities and fractional anisotropy of the connections, across populations of any size. This allows for visual and statistical comparison between groups such as males and females,[14] differing age cohorts, or healthy controls and patients. Some versions have been used to analyze how partitioned networks are in patient populations[15] or the relative balance between inter- and intra-hemispheric connections.[16]

There are many possibilities for which measures are included in the rings of a connectogram. Irimia and Van Horn (2012) have published connectograms which examine the correlative relationships between regions and uses the figures to compare the approaches of graph theory and connectomics.[17] Some have been published without the inner circles of cortical metrics.[18] Others include additional measures relating to neural networks,[19] which can be added as additional rings to the inside to show metrics of graph theory, as in the extended connectogram here:

A connectogram of a healthy control subject, and includes 5 additional nodal measures not included in the standard connectogram. From outside to inside, the rings represent the cortical region, grey matter volume, surface area, cortical thickness, curvature, degree of connectivity, node strength, betweenness centrality, eccentricity, nodal efficiency, and eigenvector centrality. Between degree of connectivity and node strength, a blank ring has been added as a placeholder.