A phase based method for investigating the functional connectivity in the fMRI data

Abstract

Detection of functional MRI activation signal and observing the functional connectivity in the spatial domain are two important aspects in the analysis of functional MRI data. There are various approaches to test for functional connectivity. One of these is based on the correlation analysis in time domain. However, it assumes that the haemodynamic response model be a linear system. On the other hand this assumption may not always be true. Therefore, a different approach towards the signal detection and spatial functional connectivity is presented in this paper. Our method is based on the phases of the time series data. Concept of phase difference vector norm has been implemented. Results are presented on a clinical data set obtained from the ringer tapping experiment. Clinical relevance of functional connectivity is described along with the background information.