Journal: bioRxiv

Article Title: Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

doi: 10.1101/2024.11.27.625681

Figure Lengend Snippet: 105 ICNs from NeuroMark 2.2 template (NeuroMark_fMRI_2.2_modelorder-multi). Presented as composite maps of 7 domains and 14 subdomains. Cerebellar Domain, Visual Domain (includes Occipitotemporal and Occipital subdomains), Paralimbic Domain, Subcortical Domain (includes Extended Hippocampal, Extended Thalamic and Basal Ganglia subdomains), Sensorimotor Domain, Higher Cognition Domain (includes Insular Temporal, Temporoparietal and Frontal subdomains) and Triple Network Domain (includes Central Executive, Default Mode, and Salience subdomains).

Article Snippet: To enhance the generalizability and reliability of findings across studies, researchers have recently delineated 105 multi-scale ICNs (NeuroMark_fMRI_2.2_modelorder-multi) from a large dataset of over 100k rsfMRI subjects , which can serve as a robust template for comparative analysis across multiple datasets and studies ( ; ).

Techniques:

Journal: bioRxiv

Article Title: Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

doi: 10.1101/2024.11.27.625681

Figure Lengend Snippet: A pipeline for Burst independent component analysis (bICA), a blind approach involving estimation of Independent Components (ICs), ICN selection, and back-reconstruction. The method initiates with subject-level PCA, followed by group-level PCA and Infomax optimization, iterated across multiple model orders to generate a comprehensive set of multi-scale Independent Components (ICs), from which the top 105 ICs were selected based on Pearson’s correlation with the NeuroMark 2.2 template. These ICNs served as references for back-reconstruction, specifically MOO-ICAR, to generate subject-level spatial maps and time courses. Subsequently, the time courses were used to generate Functional Network Connectivity (FNC).

Article Snippet: To enhance the generalizability and reliability of findings across studies, researchers have recently delineated 105 multi-scale ICNs (NeuroMark_fMRI_2.2_modelorder-multi) from a large dataset of over 100k rsfMRI subjects , which can serve as a robust template for comparative analysis across multiple datasets and studies ( ; ).

Techniques: Selection, Functional Assay

Journal: bioRxiv

Article Title: Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

doi: 10.1101/2024.11.27.625681

Figure Lengend Snippet: Best Matched 105 ICNs estimated using group-level bICA from the infant dataset. Presented as composite maps of 7 domains and 14 subdomains. Cerebellar Domain, Visual Domain (includes Occipitotemporal and Occipital subdomains), Paralimbic Domain, Subcortical Domain (includes Extended Hippocampal, Extended Thalamic and Basal Ganglia subdomains), Sensorimotor Domain, Higher Cognition Domain (includes Insular Temporal, Temporoparietal and Frontal subdomains) and Triple Network Domain (includes Central Executive, Default Mode, and Salience subdomains).

Article Snippet: To enhance the generalizability and reliability of findings across studies, researchers have recently delineated 105 multi-scale ICNs (NeuroMark_fMRI_2.2_modelorder-multi) from a large dataset of over 100k rsfMRI subjects , which can serve as a robust template for comparative analysis across multiple datasets and studies ( ; ).

Techniques:

Journal: bioRxiv

Article Title: Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

doi: 10.1101/2024.11.27.625681

Figure Lengend Snippet: Analysis of 105 ICNs estimated from the Infant Dataset from bICA and matched with the NeuroMark 2.2 Template a) Spatial similarity (rho values) for each of the 105 best-matched group-level ICNs estimated using bICA, compared with the NeuroMark 2.2 template. This panel illustrates the individual correlation values indicating how well each ICN aligns with the template b) Distribution of spatial correlation (rho values) with the template for the 105 ICNs from the infant dataset. This histogram shows the range and frequency of similarity values, providing an overview of the alignment between the estimated ICNs and the template. c) Average Functional Network Connectivity (FNC) between the ICNs derived from the infant dataset. This panel presents the mean connectivity patterns among the identified ICNs, offering insights into their functional relationships. d) Comparison of spatial maps for ICN 64 from different datasets: NeuroMark 2.2 template, the infant dataset, and a null dataset. This example highlights the similarities and differences in ICNs obtained from the actual data versus the null dataset. e) Subject-level ICNs reconstructed using the 105 best-matched ICNs. This panel shows the correlation with the template for subject-level ICNs obtained using bICA from both the infant and null datasets, demonstrating the reliability of the ICNs in the infant dataset.

Article Snippet: To enhance the generalizability and reliability of findings across studies, researchers have recently delineated 105 multi-scale ICNs (NeuroMark_fMRI_2.2_modelorder-multi) from a large dataset of over 100k rsfMRI subjects , which can serve as a robust template for comparative analysis across multiple datasets and studies ( ; ).

Techniques: Functional Assay, Derivative Assay, Comparison

Journal: bioRxiv

Article Title: Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

doi: 10.1101/2024.11.27.625681

Figure Lengend Snippet: Average 105 ICNs estimated using subject-level MOO-ICAR (NeuroMark Framework) from the infant dataset. Presented as composite maps of 7 domains and 14 subdomains. Cerebellar Domain, Visual Domain (includes Occipitotemporal and Occipital subdomains), Paralimbic Domain, Subcortical Domain (includes Extended Hippocampal, Extended Thalamic and Basal Ganglia subdomains), Sensorimotor Domain, Higher Cognition Domain (includes Insular Temporal, Temporoparietal and Frontal subdomains) and Triple Network Domain (includes Central Executive, Default Mode, and Salience subdomains).

Article Snippet: To enhance the generalizability and reliability of findings across studies, researchers have recently delineated 105 multi-scale ICNs (NeuroMark_fMRI_2.2_modelorder-multi) from a large dataset of over 100k rsfMRI subjects , which can serve as a robust template for comparative analysis across multiple datasets and studies ( ; ).

Techniques:

Journal: bioRxiv

Article Title: Born Connected: Do Infants Already Have Adult-Like Multi-Scale Connectivity Networks?

doi: 10.1101/2024.11.27.625681

Figure Lengend Snippet: Analysis of 105 ICNs Estimated Using the NeuroMark Framework from the Infant Dataset a) Spatial similarity between the average estimated subject-level ICNs and the NeuroMark 2.2 template. This panel shows the mean correlation values for ICNs across subjects, highlighting their alignment with the template. b) Distribution of spatial correlation between the 105 ICNs from the infant dataset and the template. This histogram illustrates the range and frequency of similarity values, providing an overview of how well the estimated ICNs match the template. c) Average Functional Network Connectivity (FNC) among the ICNs estimated using NeuroMark Framework. This panel presents the mean connectivity patterns between the identified ICNs, offering insights into their functional relationships d) Comparison of spatial maps for ICN 64 across different datasets: NeuroMark 2.2 template, the infant dataset, and the null dataset. e) Correlation with the template for each subject-level ICN obtained using NeuroMark Framework from both the infant and null datasets.

Article Snippet: To enhance the generalizability and reliability of findings across studies, researchers have recently delineated 105 multi-scale ICNs (NeuroMark_fMRI_2.2_modelorder-multi) from a large dataset of over 100k rsfMRI subjects , which can serve as a robust template for comparative analysis across multiple datasets and studies ( ; ).

Techniques: Functional Assay, Comparison