Pattern Analysis of Brain Functional Connectivity Parameters After Removal of Artifactual Motifs from EEG During Meditation

Abstract

The actual picture of brain network and their functional connectivity analysis is a challenging task due to large data dimensionality and inherent dynamics as well as nonlinear property of electroencephalogram (EEG). This nonlinearity and nonstationarity are sometimes introduced in EEG, because of various physiologic and non-physiologic artifacts. To overcome this, the noninvasive EEG data are used to find the functional brain network in missing data of EEG which are obtained after removal of the artifactual motifs. This paper deals with the problem of extracting the functional brain connectivity in missing EEG data that can be used for both data analysis and the classification problem such as brain–computer interface (BCI) implication. Three significant parameters, namely, transitivity, characteristics path length, and centrality are considered from the list of brain functional connectivity metrics. The comparison has been performed between continuous and discontinuous (artifactual motifs removed) data. The results have been shown that the missing data’s brain functional connectivity metrics are also following the same pattern of continuous data. The results show the implication of the framework in different brain regions involving in specific relaxation technique of short Kriya Yoga meditation.