Meta menu:

From here, you can access the Emergencies page, Contact Us page, Accessibility Settings, Language Selection, and Search page.

Open Menu

Network Neurosciences and Machine Learning

You are here:

Current Team

Head: Dr. rer. medic. J. Kruschwitz, habil., Dipl.-Psych. (CCM)
Members: Johann Kruschwitz, Emin Serin, Lea Waller, Justin Böhmer, Pablo Reinhard, Dominik Gölller 

About the group:
Brain network theory and connectomics are steadily gaining momentum in the evaluation of the whole brain as an interconnected network, emerging as a powerful tool for capturing the complexity of the brain's function and structure. With connectomics, the structural and functional architecture of the human brain can be studied by defining networks, which comprise regions of interests (“nodes”) and interregional structural or functional connections (“edges”). In this context, graph theory, the mathematical study of networks, provides a powerful and comprehensive formalism of global and local network properties of complex structural or functional brain connectivity. Application of graph theoretical measures to clinical populations has revealed differences in these properties in many neurological and psychiatric disorders. Apart from describing topological network properties, graph theory also provides a framework for identification of anatomically localized sub-networks associated with particular effects of interest (such as candidate genotype group differences or correlations with neuropsychological test scores) across the entire brain. Given their rich, multiscale, and high-dimensional feature space, graph models of the brain (i.e. the connectome) also herald promising opportunities in search of neural biomarkers. Here, the application of machine-learning and deep-learning, i.e. decoding, models allows for a pattern-oriented exploration of the connectome, offering a promising solution to the identification of previously overlooked mechanisms of brain function. 

In our group we apply computational techniques from the field of connectomics and machine-learning to derive rigorous neuroscientific conclusions across a wide range of topics (see current projects). Moreover, we work on the development of tools and techniques to investigate and describe the brains intrinsic architecture (Connectomics methods: GraphVar & NBS-Predict).