A brain connectivity metric based on phase linearity measurement

The analysis of brain connectivity is gaining interest in recent years due to the relevant information it carries about the functioning of the brain in health and in disease. In brief, it consists in measuring the statistical dependencies between signals generated by different brain regions. Several metrics have been proposed in literature, related to three families: amplitude based, phase based on jointly amplitude and phase based. Due to the large amount of noise that typically affects the estimation of the connectivity maps, averaging over several epochs of a population is normally carried out. We propose a novel phase based metric, namely the Phase Linearity Metric (PLM), that is resilient to noise and volume conduction, bearing promise to lower the number of epochs needed for a reliable measurement. The comparison with the widely adopted PLI connectivity metric confirms the effectiveness of the PLM.