This talk will discuss how approaches such as manifold learning can help understanding individuality of the human brain architecture, the reorganization of functional networks during disease, and the development of the human brain. I will illustrate the clinical relevance of insights gained by these methods, and how they may inform treatment decisions. The second part of the talk will explain how we can use methods from geometry to reconstruct 77 million years of brain evolution based on data from 90 living species. After discussing the methodology and its underlying assumptions, the talk will provide insights gained from this model, including the association of brain areas and behavior, and the possible link between heterogeneous patterns of cortical evolution and the emergence of human cognitive capabilities.
Data, paper, code and video for brain evolution (also follow link to github from that page): www.cir.meduniwien.ac.at/evolution
Brain development: www.cir.meduniwien.ac.at/fetal