Applications of granger causality model to connectivity network based on fMRI time series

  • Authors:

  • Xiao-Tong Wen;Xiao-Jie Zhao;Li Yao;Xia Wu

  • Affiliations:

  • State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China;School of Information Science and Technology, Beijing Normal University, Beijing, China;School of Information Science and Technology, Beijing Normal University, Beijing, China;State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China

  • Venue:
  • ICNC'06 Proceedings of the Second international conference on Advances in Natural Computation - Volume Part I
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The connectivity network with direction of brain is a significant work to reveal interaction and coordination between different brain areas. Because Granger causality model can explore causal relationship between time series, the direction of the network can be specified when the model is applied to connectivity network of brain. Although the model has been used in EEG time sires more and more, it was seldom used in fMRI time series because of lower time resolution of fMRI time series. In this paper, we introduced a pre-processing method to fMRI time series in order to alleviate the magnetic disturbance, and then expand the time series to fit the requirement of time-variant algorism. We applied recursive least square (RLS) algorithm to estimate time-variant parameters of Granger model, and introduced a time-variant index to describe the directional connectivity network in a typical finger tapping fMRI experiment. The results showed there were strong directional connectivity between the activated motor areas and gave a possibility to explain them.