Functional brain network properties of healthy full-term newborns quantified by scalp and source-reconstructed EEG. (March 2023)

• Record Type: Journal Article
• Authors: Chirumamilla, Venkata C; Hitchings, Laura; Mulkey, Sarah B.; Anwar, Tayyba; Baker, Robin; Larry Maxwell, G.; De Asis-Cruz, Josepheen; Kapse, Kushal; Limperopoulos, Catherine; du Plessis, Adre; Govindan, R.B.
• Is Part Of: Clinical neurophysiology. Volume 147(2023)
• Page Start: 72
• Page End: 80
• Publication Date: 2023-03
• Subjects: Source analysis -- Linearly constrained minimum variance beamformer -- Graph theory -- Coherence analysis -- Low-risk term newborns; EEG Electroencephalogram -- fMRI Functional Magnetic Resonance Imaging -- BOLD Blood Oxygen Level Dependent -- MEG Magnetoencephalogram -- GA Gestational Age -- HD-EEG High-Density EEG -- EKG Electrocardiogram -- Apgar appearance, pulse, grimace, activity, and respiration -- ROC Receiver Operating Characteristic Curve -- AUC Area Under the Curve -- MRI Magnetic Resonance Imaging -- MNI Montreal Neurological Institute -- BET Brain Extraction Tool -- FSL FMRIB Software Library -- BEM Boundary Element Method -- AAL Automated Anatomical Labeling -- ROI Regions of Interest -- SWP Small-World Propensity -- AIC Akaike Information Criterion; Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2023.01.005
• Languages: English
• ISSNs: 1388-2457

Abstract

Highlights: Functional connectivity of term low-risk newborns was characterized using scalp and source EEG. Source-based networks showed the presence of small-world architecture with four modules connected by hubs and rich-club topology. Source analysis of HD-EEG may be helpful to study brain functional connectivity at the bedside of critically ill infants. Abstract: Objective: Identifying the functional brain network properties of term low-risk newborns using high-density EEG (HD-EEG) and comparing these properties with those of established functional magnetic resonance image (fMRI) – based networks. Methods: HD-EEG was collected from 113 low-risk term newborns before delivery hospital discharge and within 72 hours of birth. Functional brain networks were reconstructed using coherence at the scalp and source levels in delta, theta, alpha, beta, and gamma frequency bands. These networks were characterized for the global and local network architecture. Results: Source-level networks in all the frequency bands identified the presence of the efficient small world (small-world propensity (SWP) > 0.6) architecture with four distinct modules linked by hub regions and rich-club (coefficient > 1) topology. The modular regions included primary, association, limbic, paralimbic, and subcortical regions, which have been demonstrated in fMRI studies. In contrast, scalp-level networks did not display consistent small world architecture (SWP < 0.6), and also identified only 2–3 modules in each frequency band. The modular regions of the scalp-network primarily included frontal and occipital regions. Conclusions: Our findings show that EEG sources in low-risk newborns corroborate fMRI-based connectivity results. Significance: EEG source analysis characterizes functional connectivity at the bedside of low-risk newborn infants soon after birth.