Abstract

Neuroinformatics: From Big Data to Clinical Applications

Mechanistic models of neural networks, that have been the focus neuroinformatics research in the past decades, have uncovered important organizational principles of the brain. The exponential rise in the availability of neural data now enables us to move beyond hand-crafted mechanistic models and develop machine learning algorithms that uncover novel organizational principles in a data-driven fashion. In the first part of my presentation, I introduce machine learning algorithms developed in my team that enable novel insights into causal relations between neural activity and cognitive variables. In the second part of my talk, I illustrate how these methods allow us to translate basic research into clinical applications in the domains of stroke rehabilitation and brain-computer interfacing. I conclude by arguing that computer science in general and neuroinformatics in particular has the unique opportunity to shape future research in cognitive- and translational neuroscience.

Bio

Moritz Grosse-Wentrup is Professor for Data Science at the Ludwig-Maximilians-Universität München and group leader at the Max Planck Institute for Intelligent Systems, Tübingen. After obtaining his Dr.-Ing. degree at Technische Universität München, he was a postdoc with Bernhard Schölkopf at the Max Planck Institute for Biological Cybernetics. He has been the recipient of the 2011 Annual BCI Research Award and the 2016 IEEE Brain Initiative Best Paper Award. He serves as the steering committee chair for the International Workshop on Pattern Recognition in Neuroimaging (PRNI) and as an area chair for the Neural Information Processing Systems (NIPS) conference. He is a founding member of the EURASIP-SAT on Biomedical Image and Signal Analysis. His research focuses on machine learning for brain decoding and neural engineering, with applications in brain-computer interfacing for communication and rehabilitation.