A New Method for Estimating Spectral Clustering Change-Points for fMRI Data

• Activity Number: 89
• Type: Invited
• Date/Time: Sunday, August 9, 2015 : 8:30 PM to 9:15 PM
• Sponsor: Korean International Statistical Society
• Author(s): Ivor Cribben* and Yi Yu
• Companies: University of Alberta and University of Cambridge
• Keywords: Change point analysis ; Spectral clustering ; Community detection ; Functional connectivity ; Brain networks ; fMRI

Abstract

Spectral clustering is a computationally feasible and model-free method widely used in the identification of communities in networks. In this work, we introduce a data-driven method that detects change points in the network structure of a multivariate time series, with each component of the time series represented by a node in the network. Spectral clustering allows us to consider high dimensional time series where the number of time series is greater than the number of time points (n< p). The method allows for estimation of both the time of change in the network structure and the graph between each pair of change points, without prior knowledge of the number or location of the change points. Permutation and bootstrapping methods are used to perform inference on the change points. The method is applied to various simulated high dimensional data sets as well as to a resting state functional magnetic resonance imaging (fMRI) data set. The results illustrate the method's ability to observe how the network structure between different brain regions changes over the experimental time course. The method promises to offer a deep insight into the inner workings and dynamics of the brain.