Abstract
Brain activity over the entire network is complex. A full understanding of brain activity requires careful study of its multi-scale spatial-temporal organization. Motivated by these challenges, we will explore some characterizations of dependence between components of a multivariate time series and then apply these to the study of brain functional connectivity. This is potentially interesting for brain scientists because functional brain networks are associated with cognitive function, and mental and neurological diseases. There is no single measure of dependence that can capture all facets of brain connectivity. In this talk, we shall present some new models for exploring potential interactions between oscillations in multivariate time series including non-linear (e.g., phase-amplitude relationships). The proposed approach captures lead-lag relationships and hence can be used as a general framework for spectral causality. We will also demonstrate some inferential tools for assessing Granger causality in brain functional networks.
This is joint work with Marco Pinto (UC Irvine) and some members of the Biostatistics Group: Paolo Redondo and Malik Shahid.
Brief Biography
Hernando Ombao is Professor of Statistics at KAUST. He is the PI of the Biostatistics Research Group. His research is on statistical models for time series with dynamic and complex structures. Prior to coming to KAUST, he was a faculty member at UC Irvine, Brown University, University of Illinois and the University of Pittsburgh. At UC-Irvine, he was a recipient of the Mid-Career Distinguished Research Award. He is Co-Editor of the Handbook of Statistical Methods for Neuroimaging. He was PI of several US NSF awards. His service to the community includes roles as AE of Statistics journals (Metron, JASA, JRSS-B, CSDA) and as a permanent member of the NIH Biostatistics Study Section. He is a founding member of the Statistics in Imaging Section and was elected as chair in 2020. He is an Elected Fellow of American Statistical Association.