On the Effect of Volume Conduction on Graph Theoretic Measures of Brain Networks in Epilepsy

Abstract

It is well established that both volume conduction and the choice of recording reference (montage) affect the connectivity measures obtained from scalp EEG, in the time and frequency domains. A number of measures have been proposed aiming to reduce this influence. Our purpose in this work is to establish the extent to which volume conduction and montage influence the graph theoretic measures of brain networks in epilepsy obtained from scalp EEG. We evaluate and compare two standard and most commonly used linear connectivity measures—cross-correlation in the time domain and coherence in the frequency domain—with measures that account for volume conduction, namely, corrected cross-correlation, imaginary coherence, phase lag index, and weighted phase lag index. We show that the graphs constructed with cross-correlation and coherence are affected by volume conduction and montage more markedly; however, they demonstrate the same trend—decreasing connectivity at seizure onset, which continues decreasing in the ictal and early postictal period, increasing again several minutes after the seizure has ended—with all other measures except imaginary coherence. In particular, networks constructed using cross-correlation yield better discrimination between the pre-ictal and ictal periods than the measures less sensitive to volume conduction such as the phase lag index and imaginary coherence. Thus, somewhat paradoxically, although removing the effects of volume conduction allows for a more accurate reconstruction of the true underlying networks this may come at the cost of discrimination ability with respect to brain state.