Network Parameters for Studying Functional Connectivity in Brain MEG Data

The functional connectivity of various brain regions has been studied here using the knowledge from two different scientific fields. The methods of Synchronization Likelihood (SL) and network theory are applied to magnetoencephalography (MEG) data in an effort to study the brain as a complex network. In this paper the SL method has been used to characterize the functional interactions as ‘‘functional connectivity’’, by performing measures of statistical interdependencies between brain activity signals. The underlying assumption is that such correlations, at least in part, reflect the functional interactions between different brain regions. Methods applied in this study investigate the occurrence of small-world phenomenon in MEG data by considering the application of the SL method and the characterization of the respective graphs obtained by varying the threshold T. The data set used here is from a single subject performing a yogic breathing exercise. In the results we show how we are able to characterize and differentiate the different phases of the breathing exercise by using the index σ that defines the presence of a small world network, along with other network parameters that include the clustering coefficient, the characteristic path length, and the nodes degree.