Effect of channel density, inverse solutions and connectivity measures on EEG resting-state networks reconstruction: A simulation study

• DOI: 10.1016/j.neuroimage.2023.120006
• Publication Date: 2023-03-11T07:04:57Z
• Authors: Sahar Allouch; Aya Kabbara; Joan Duprez; Mohamad Khalil; Julien Modolo; Mahmoud Hassan
• Keywords: Brain Mapping; Channel density; Reproducibility of Results; Brain; Neurosciences. Biological psychiatry. Neuropsychiatry; Electroencephalography; Functional connectivity; Analytical variability; EEG resting-state networks; Connectome; Humans; [SDV.IB]Life Sciences [q-bio]/Bioengineering; Neural mass model; Computer Simulation; Inverse solution; RC321-571

Along with the study of brain activity evoked by external stimuli, past two decades witnessed an increased interest in characterizing spontaneous occurring during resting conditions. The identification connectivity patterns this so-called "resting-state" has been subject a great number electrophysiology-based studies, using Electro/Magneto-Encephalography (EEG/MEG) source method. However, no consensus reached yet regarding unified (if possible) analysis pipeline, and several involved parameters methods require cautious tuning. This is particularly challenging when different analytical choices induce significant discrepancies results drawn conclusions, thereby hindering reproducibility neuroimaging research. Hence, our objective was to shed light on effect variability outcome consistency evaluating implications EEG accuracy resting-state networks (RSNs) reconstruction. We simulated, neural mass models, data corresponding RSNs, namely default mode network (DMN) dorsal attentional (DAN). investigated impact five channel densities (19, 32, 64, 128, 256), three inverse solutions (weighted minimum norm estimate (wMNE), exact low-resolution electromagnetic tomography (eLORETA), linearly constrained variance (LCMV) beamforming) four functional measures (phase-locking value (PLV), phase-lag index (PLI), amplitude envelope correlation (AEC) without leakage correction), correspondence between reconstructed reference networks. showed that, related electrodes, reconstruction algorithm, measure, high present results. More specifically, show that higher channels significantly Additionally, performance tested measures. Such methodological absence standardization represent critical issue for studies should be prioritized. believe work could useful field electrophysiology connectomics, increasing awareness challenge approaches its reported