СЕМИНАР: др Босиљка Тадић

У оквиру семинара Центра за изучавање комплексних система Института за физику, у четвртак, 10. октобра 2019. у 14 часова у читаоници библиотеке „Др Драган Поповић“, проф. др Босиљка Тадић (Департман за теоријску физику института „Јожеф Стефан“ у Љубљани, и Complexity Science Hub у Бечу) одржаће предавање:

Computational Modeling of the Brain & Hidden Structure of Human
Connectomes

САЖЕТАК:

The modern science of the brain explores how the brain’s anatomical
parts and complex pathways between them govern human activity,
psychology and behaviour. A  network that maps the organization of
neuronal connections between brain regions is known as the human
connectome. Currently, a large amount of experimental data has been
collected within the Human Connectome Project (HCP) to enable different
aspects of brain research. However,  the procedure to extract the
network from the fMRI or another brain imaging methods, as well as the
analysis of whole-brain networks, requires specific computational
modelling approaches. In this seminar, we briefly review the necessary
procedures preceding the construction of a brain network and describe
some parameters that affect the outcome. Further, we discuss in detail
[1] the consensus connectomes, that we generate for a specified set of
parameters on the Budapest Conenctome Server [2] based on HCP data. By
analyzing higher-order connections, we show how the consensus
connectomes of female subjects systematically differ from the one for
male subjects. Meanwhile, their standard graph-theoretic properties
remain self-similar. These robust gender differences within the
higher-order connectivity in human brain networks imply their different
functional properties; whereas, the understanding of the origin and
interpretation of these differences remain beyond the mathematical
modelling framework.

[1] B. Tadić, M. Andjelković, R. Melnik, Functional geometry of human
connectomes, Scientific Reports 9:12060 (2019).

[2] B. Szalkai, C. Kerepesi, B. Varga, V.  Grolmusz, Parameterizable
Consensus Connectomes from the Human Connectome Project: The Budapest
Reference Connectome Server v3.0, Cognitive Neurodynamics, (2016).