Longitudinal Changes in Regional Fiber Length Map in Early School-age Children
Poster No:
1229
Submission Type:
Abstract Submission
Authors:
Yanlin Yu1, Qing Cai1, Chu-Chung Huang1
Institutions:
1East China Normal University, Shanghai, China
First Author:
Co-Author(s):
Introduction:
The understanding of brain networks is of great significance for human cognition and behavior, especially in the development stage. The non-invasive neuroimaging techniques that estimate the structural connectivity in brain white matter (WM) using diffusion MRI tractography were proposed to identify the human brain's fundamental organization. However, mapping structural connectome using diffusion MRI has been challenging due to streamline quantification bias and may impede the sensitivity of change in network metrics (Yeh et al., 2021). Recent study suggested that the fiber length distribution could be a novel way to differentiate a brain region served as the functional integrator or local modules (Bajada et al., 2019). Here, we aim to generate a regional index characterized by length distribution of specific regions to investigate the developmental change during early school age. We hypothesized that longitudinal changes in regional fiber length distribution provide insight into how brain regions modulate their communication cost and may mirror the network hierarchy's developmental process.
Methods:
A total of 30 typically developing children were included in this study (7.5 ± 0.3 yr, 18 females and 12 males) at the first year in primary school and received the follow-up scanning in 1 year (8.6 ± 0.3 yr). To control the effect of brain size on fiber length, the longitudinal group FOD (fiber orientation distribution) template was generated and each individual's FOD was transformed into this template for whole-brain tractography (Genc et al., 2018). The individual length connectome was reconstructed for the whole-brain tractogram based on the HCPex atlas (426 brain regions) (Huang et al., 2022). The length distribution was defined as the connected length distribution of a specific node, the median length of which was then calculated to represent the distribution characteristic. We compared the length median of brain regions in two time points using paired two-sample permutation test to identify the significant developmental changes in regional length distribution (1000 permutations, p < 0.01) (Figure 1). To explain the mechanism underlying developmental changes in median length, we compared the count of edges constituting the distribution across various length range at two time points. To further investigate the topological meaning of regional length map, we examined the correlation between length median and graph metrics based on the binary structural connectome (FDR corrected, p < 0.05).
Results:
Most brain regions' length median decreased during development, the brain regions with significantly decreased in length median (termed length-decreased ROIs) includes L_TF, L_PHA1, L_TE1m, L_7Am, L_7PL, L_VIP, L_25, L_a24, R_PBelt and R_10d. Whereas, only R_FOP1 show a significant increase in length median (termed length-increased ROI). We observed a general pattern that there was an increase in short-range connections, accompanied by a decrease in long-range connections in length-decreased ROIs, and conversely, an increase in long-range connections in length-increased ROI. Additionally, the regional length map was positively correlated with degree and centrality but negatively correlated with the clustering coefficient and local efficiency (Figure 2).
Conclusions:
This study first proposed the regional length map to investigate the developmental changes in WM during early school age. For most brain areas, the length distributions decreased with age, which could be explained by the pruning of long-range WM and the growth of short-range WM connections. Moreover, the length map positively correlated with integration and negatively correlated with segregation of structural network. Our findings suggest that, compared to streamline-based structural network metrics, the length map of brain regions may serve as a potential biomarker for assessing the development of brain topological organization at a structural level.
Lifespan Development:
Early life, Adolescence, Aging 1
Normal Brain Development: Fetus to Adolescence
Modeling and Analysis Methods:
Diffusion MRI Modeling and Analysis 2
Keywords:
Development
Tractography
White Matter
Other - Regional Length Map;Topology
1|2Indicates the priority used for review
Provide references using author date format
Genc, S. (2018). Development of white matter fibre density and morphology over childhood: A longitudinal fixel-based analysis. NeuroImage, 183, 666–676. https://doi.org/10.1016/j.neuroimage.2018.08.043
Huang, C.-C. (2022). An extended Human Connectome Project multimodal parcellation atlas of the human cortex and subcortical areas. Brain Structure and Function, 227(3), 763–778. https://doi.org/10.1007/s00429-021-02421-6
Yeh, C.-H. (2021). Mapping Structural Connectivity Using Diffusion MRI: Challenges and Opportunities. Journal of Magnetic Resonance Imaging, 53(6), 1666–1682. https://doi.org/10.1002/jmri.27188