Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 63 of the 65 references it cites were found in a registry or an index; 4 of the 43 sources its claims cite were checked for support.

Functional properties of resting state networks in healthy full-term newborns

Open access
De Asis-Cruz, Josepheen, Bouyssi-Kobar, Marine, Evangelou, Iordanis, Vezina, Gilbert, Limperopoulos, Catherine · 2015 · Scientific Reports

Abstract

Objective, early, and non-invasive assessment of brain function in high-risk newborns is critical to initiate timely interventions and to minimize long-term neurodevelopmental disabilities. A prerequisite to identifying deviations from normal, however, is the availability of baseline measures of brain function derived from healthy, full-term newborns. Recent advances in functional MRI combined with graph theoretic techniques may provide important, currently unavailable, quantitative markers of normal neurodevelopment. In the current study, we describe important properties of resting state networks in 60 healthy, full-term, unsedated newborns. The neonate brain exhibited an efficient and economical small world topology: densely connected nearby regions, sparse, but well integrated, distant connections, a small world index greater than 1, and global/local efficiency greater than network cost. These networks showed a heavy-tailed degree distribution, suggesting the presence of regions that are more richly connected to others ('hubs'). These hubs, identified using degree and betweenness centrality measures, show a more mature hub organization than previously reported. Targeted attacks on hubs show that neonate networks are more resilient than simulated scale-free networks. Networks fragmented faster and global efficiency decreased faster when betweenness, as opposed to degree, hubs were attacked suggesting a more influential role of betweenness hub in the neonate network.

Overview

CoChat Summary

## TL;DR

This study used resting-state fMRI and graph theory in 60 healthy, full-term, unsedated newborns to characterize normative neonatal functional network topology. The neonatal brain showed small-world, cost-efficient, modular, and resilient networks, with targeted attacks on betweenness hubs causing greater efficiency loss than attacks on degree hubs.

## Key findings

  • In 60 healthy, full-term newborns (median GA at birth 39.64 ± 0.74 weeks; median age at scan 12.5 ± 6 days), neonate functional networks met small-world criteria: at all thresholds tested, the small world index was significantly greater than 1, with SWI ranging from 1.15 to 2.70.
  • Network economy was observed: at network costs from 0 to 0.6, global and local efficiency exceeded network cost; maximal cost efficiency was 0.3437 at cost = 0.22, when edge density was 22% (881 of 4005 possible edges).
  • The degree distribution was heavy-tailed and best fit by an exponentially truncated power law, not a pure power law or exponential distribution.
  • The neonate brain showed strong modular organization with modularity Q = 0.6447 and four modules; degree hubs were mainly in Modules 3, while most betweenness hubs were in Module 1.
  • After 20% of nodes were removed in targeted-attack simulations, 80% and 74% of nodes remained in the largest connected group when degree and betweenness hubs were targeted, respectively; global efficiency retained 48.1%–56.7% of the original network’s efficiency. Isolated removal of betweenness hubs caused a greater drop in global efficiency than removal of degree hubs (p < 0.01).

## Methods

  • Observational normative study of 72 recruited full-term healthy neonates; 60 were included in the final analyses after 12 were excluded for failing motion criteria. All infants had structurally normal brains on neuroradiologist review.
  • Resting-state fMRI was acquired on a 3T scanner while infants slept unsedated; preprocessing included motion censoring/scrubbing, nuisance regression, band-pass filtering (0.01 < f < 0.1), and parcellation into 90 ROIs based on the AAL atlas mapped to neonates.
  • Graph analysis used 90-node functional connectivity matrices thresholded across correlation and degree ranges; metrics included clustering coefficient, path length, small-world index, global/local efficiency, hub detection by degree and betweenness centrality, Louvain modularity detection, and resilience analyses via random failure and targeted hub removal compared with random and scale-free networks.

## Limitations

  • The paper states it could not evaluate potential sources of individual differences, including gender, socio-economic status, and mode of delivery.
  • The authors note they used representative hub metrics only (degree and betweenness) and that other measures, such as eigenvector, closeness, leverage centrality, and participation coefficient, were outside the scope of the paper.
  • Evident from the text: the study used a coarse parcellation of about 90 nodes, which limits spatial resolution compared with voxel-based approaches; the paper explicitly notes this as a general limitation of the analysis.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/srep17755

Cites

63 of 63 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Small-world properties of neonate resting state networks as a function of correlation threshold R. (A) Group-averaged clustering coefficient for neonates, Cneo (black), and random networks, Crandom (red). Cneo is significantly greater than Crandom at all thresholds evaluated, p < 0.01, Bonferroni corrected (green asterisks). (B) Group-averaged characteristic path length for neonates, Lneo (black), and random networks, Lrandom (red). Lnet and Lrandom are not different (blue asterisks) at threshold range 0 < R < 0.175. Degree, k, and degree distribution, p(k), of neonate networks are preserved in random networks. (C) Small world index, SWI or σ (light blue). SWI > 1 (99% CI) for all thresholds tested. Also shown are normalized clustering coefficient (γ, black) and normalized characteristic path length (λ, red). Here, γ >> 1 and λ ≥ 1. Error bars are standard error of the mean (SEM).
Figure 1 Small-world properties of neonate resting state networks as a function of correlation threshold R. (A) Group-averaged clustering coefficient for neonates, Cneo (black), and random networks, Crandom (red). Cneo is significantly greater than Crandom at all thresholds evaluated, p < 0.01, Bonferroni corrected (green asterisks). (B) Group-averaged characteristic path length for neonates, Lneo (black), and random networks, Lrandom (red). Lnet and Lrandom are not different (blue asterisks) at threshold range 0 < R < 0.175. Degree, k, and degree distribution, p(k), of neonate networks are preserved in random networks. (C) Small world index, SWI or σ (light blue). SWI > 1 (99% CI) for all thresholds tested. Also shown are normalized clustering coefficient (γ, black) and normalized characteristic path length (λ, red). Here, γ >> 1 and λ ≥ 1. Error bars are standard error of the mean (SEM).CC BY · PubMed Central
Figure 2 Economy of neonate small world networks. Global (A, black; Eglob) and local (B, black; Eloc) efficiency are greater than network cost (Costneo). Neonate RSNs are cost efficient; Eglob – cost is always positive (C). Efficiency of neonate networks lie between random (red) and lattice/regular (green) networks.
Figure 2 Economy of neonate small world networks. Global (A, black; Eglob) and local (B, black; Eloc) efficiency are greater than network cost (Costneo). Neonate RSNs are cost efficient; Eglob – cost is always positive (C). Efficiency of neonate networks lie between random (red) and lattice/regular (green) networks.CC BY · PubMed Central
Figure 3 Heavy tailed degree distribution of neonate brains. Right-tailed degree distribution, p(k), of group-averaged neonate networks thresholded at R 0.25 (A), 0.35 (B) and 0.45 (C) is shown. This pattern becomes more apparent as the threshold increases and the network gets sparser. (A–C), left, shows histograms of degree distributions; right panels show log-log plots of cumulative distribution versus degree. Best fit for data (+) is an exponentially truncated power law (··), compared to power law (––) and exponential law (--).
Figure 3 Heavy tailed degree distribution of neonate brains. Right-tailed degree distribution, p(k), of group-averaged neonate networks thresholded at R 0.25 (A), 0.35 (B) and 0.45 (C) is shown. This pattern becomes more apparent as the threshold increases and the network gets sparser. (A–C), left, shows histograms of degree distributions; right panels show log-log plots of cumulative distribution versus degree. Best fit for data (+) is an exponentially truncated power law (··), compared to power law (––) and exponential law (--).CC BY · PubMed Central
Figure 4 Degree and betweenness hubs. Axial plane shows hubs based on degree (A) and betweenness (B) as big circles with bold black outlines. Broken lines divide the brain into left and right hemispheres; colors designate modules. Black lines between nodes indicate inter-module connections; intra-module connections have the same color as the nodes they connect. Majority of degree hubs belong to the limbic-paralimbic-subcortical region. Betweenness hubs mostly belong to Module 1 but are relatively more distributed in the brain.
Figure 4 Degree and betweenness hubs. Axial plane shows hubs based on degree (A) and betweenness (B) as big circles with bold black outlines. Broken lines divide the brain into left and right hemispheres; colors designate modules. Black lines between nodes indicate inter-module connections; intra-module connections have the same color as the nodes they connect. Majority of degree hubs belong to the limbic-paralimbic-subcortical region. Betweenness hubs mostly belong to Module 1 but are relatively more distributed in the brain.CC BY · PubMed Central
Figure 5 Modular organization of the neonate brain. The Louvain algorithm was used to identify modules. (A) shows the unthresholded, averaged connectivity matrix for 60 neonates and (B) shows the nodes arranged by module, and by decreasing degree within each module. (C) shows the four functional modules identified in the neonate brain. The network is shown using a Kamada-Kawai layout as implemented in the R package igraph; the graph distance between two nodes determines their location in the layout. Node sizes reflect the degree, k, of each node. Colors reflect module membership. Classification of node in the modules following Mesulam60 is shown in (D).
Figure 5 Modular organization of the neonate brain. The Louvain algorithm was used to identify modules. (A) shows the unthresholded, averaged connectivity matrix for 60 neonates and (B) shows the nodes arranged by module, and by decreasing degree within each module. (C) shows the four functional modules identified in the neonate brain. The network is shown using a Kamada-Kawai layout as implemented in the R package igraph; the graph distance between two nodes determines their location in the layout. Node sizes reflect the degree, k, of each node. Colors reflect module membership. Classification of node in the modules following Mesulam60 is shown in (D).CC BY · PubMed Central
Figure 6 Resilience of resting state networks. Response to targeted removal of degree (red) and betweenness (green) hubs and random removal of nodes (black) are shown for neonate and comparable random and scale-free networks. The effects of node removal to the largest connected cluster (A) and global efficiency (B) of the network are shown. The neonate brain is more resilient to targeted attack of hubs compared to a scale-free network. Attack on betweenness hubs in the neonate brain decreases global efficiency and the size of the largest cluster faster than attack on degree hubs.
Figure 6 Resilience of resting state networks. Response to targeted removal of degree (red) and betweenness (green) hubs and random removal of nodes (black) are shown for neonate and comparable random and scale-free networks. The effects of node removal to the largest connected cluster (A) and global efficiency (B) of the network are shown. The neonate brain is more resilient to targeted attack of hubs compared to a scale-free network. Attack on betweenness hubs in the neonate brain decreases global efficiency and the size of the largest cluster faster than attack on degree hubs.CC BY · PubMed Central
Figure 7 Isolated removal of betweenness and degree hubs. Isolated removal of betweenness hubs cause greater decreases in global efficiency compared to degree hubs (p < 0.01). (A) shows comparison of hubs (mean ± 1SD) and (B) show the top 20% of nodes (n = 18 for both groups).
Figure 7 Isolated removal of betweenness and degree hubs. Isolated removal of betweenness hubs cause greater decreases in global efficiency compared to degree hubs (p < 0.01). (A) shows comparison of hubs (mean ± 1SD) and (B) show the top 20% of nodes (n = 18 for both groups).CC BY · PubMed Central

Verification

Claims the paper makes

4 Valid · 3 Not valid · 36 Not checked · 89 Own results and interpretations

77 more claims

Create a free account to see all 132 claims and to verify the 39 references still open.

See all claims — free account
  • Probably not a substantial claim. 32
  • It may state two separate claims as one. 24
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 65 references

Cited by

See all 87

Metadata

Authors
De Asis-Cruz, J., Bouyssi-Kobar, M., Evangelou, I., Vezina, G., & Limperopoulos, C.
Affiliations
Children's National (US)
Corresponding author
Limperopoulos, Catherine
Venue
Scientific Reports
Published
2015-12-07
Volume
5
Issue
1
Pages
17755-17755
ISSN
2045-2322 (ISSN-L)
OpenAlex
W2212279317
Cited by
85
References
65
Citations by year
2026: 12 · 2025: 9 · 2024: 15 · 2023: 8 · 2022: 6 · 2021: 8 · 2020: 6 · 2019: 6 · 2018: 7 · 2017: 8
FWCI
3.01
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Funders
  • Canadian Institutes of Health Research (R01 HL116585-01, 81116_1)
  • National Heart, Lung, and Blood Institute (R01HL116585)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Functional Brain Connectivity Studies
  • Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Complex Network Analysis Techniques
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced Neuroimaging Techniques and Applications
Keywords
resting-state networks, neonatal brain, term infants, small-world networks, network hubs, betweenness centrality, degree centrality, network resilience, global efficiency, network cost, functional MRI, graph theory, brain development, high-risk infants
MeSH terms
Female, Health, Humans, Infant, Newborn, Male, Nerve Net, Rest

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Functional properties of resting state networks in healthy full-term newborns. https://app.cochat.ai/papers/10.1038/srep17755
The paper
De Asis-Cruz, J., Bouyssi-Kobar, M., Evangelou, I., Vezina, G., & Limperopoulos, C. (2015). Functional properties of resting state networks in healthy full-term newborns. Scientific Reports. https://doi.org/10.1038/srep17755

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.