Title: Discrepancy of correspondence, tracking the changing and developing functional connectome
Abstract: How can we capture the functional connectome if it is not tightly coupled to anatomical reference frames? Machine learning approaches allow for the tracking of changes in the connectome, the matching of network structures between healthy controls, and patients after a reorganization, or the detection of subtle anomalies. In the talk, we will discuss examples ranging from finding homologous areas in different species, to early brain development, and tracking reorganization in patients suffering from cancer or epilepsy. After outlining methodology, we will discuss which questions we can ask once we are able to define multiple and possibly contradictory notions of correspondences across individuals or groups.