Print Close

Normative Brain Development of Cortical Thickness in Early-School Age Children: A Longitudinal Study

Poster No:

1240

Submission Type:

Abstract Submission

Authors:

Baitao Li¹, Xiao Wang¹, Qing Cai¹, Chu-Chung Huang¹

Institutions:

¹East China Normal University, Shanghai, China

First Author:

Baitao Li
East China Normal University
Shanghai, China

Co-Author(s):

Xiao Wang
East China Normal University
Shanghai, China

Qing Cai
East China Normal University
Shanghai, China

Chu-Chung Huang
East China Normal University
Shanghai, China

Introduction:

Normative modeling has emerged as an essential analytical tool in neuroimaging, providing a framework to delineating individual difference, especially in brain development (Marquand et al., 2016). However, most previous studies relied solely on cross-sectional data to model the age-related brain changes, potentially leading to an incomplete understanding of the variability present in longitudinal data (Di Biase et al., 2023). To address this limitation of cross-sectional normative modeling, our research shifts focus to the use of longitudinal data for modeling developmental changes. In this study, we integrated the gamlssNP method (Stasinopoulos et al., 2017), specifically tailored for longitudinal data analysis. Considering the considerable environmental change experienced by early-school-age children, we model a longitudinal cohort of children aged 7 to 10 years, with at least two MRI data points. This study aims to establish a less-biased normative model based on longitudinal data to depict cortical thickness development in early-school-age children.

Methods:

We selected participants with at least two data time points, which yielded 529 T1-weighted (T1w) MRI images. These T1w images were processed using FreeSurfer (v7.3.2), to extract cortical thickness in various brain regions using DKT atlas of each participant. We fitted normative centile curves were fit to cortical thickness using gamlssNP(gamlss.mx) method in R version 4.3.0 (2023-04-21). For model selection and diagnostic, we employed 5-fold cross-validation to determine the optimal parameter combination of degrees of freedom (df) and the number of components (K) for gamlssNP model. The final model parameters were chosen to achieve a balance between fit (lower AIC) and generalization (lower VGD) across all brain regions.

Results:

After a 5-fold cross-validation, the (df = 2, K = 2) parameter combination emerges as notably effective among all tested configurations, with an VGD percentile of 0.1461 ± 0.0255, suggesting the model robustness and consistency. Normative model analysis revealed significant age correlations in several brain regions. Among the regions, precentral, paracentral, and entorhinal, exhibited a positive correlation, whereas 14 regions, including the insula, rostral anterior cingulate, and medial orbitofrontal, showed a negative correlation with age (Fig. 1). Here we showed the most significant six regions for demonstration.

Conclusions:

Our study found that the precentral, paracentral, and entorhinal cortices, increase in cortical thickness with age. This increase may imply development in motor and cognitive abilities in children. In contrast, the insula and rostral anterior cingulate show reduced thickness, suggesting neural maturation (Gilmore et al., 2018). Overall, these patterns reveal a complex developmental timeline, wherein some regions mature early, playing roles in emotional and cognitive functions. Future work should investigate the individual differences between actual and predicted cortical thickness in these regions and associate them with cognitive development, especially in early school-age children.

Lifespan Development:

Early life, Adolescence, Aging ¹

Modeling and Analysis Methods:

Classification and Predictive Modeling ²

Keywords:

Development

Modeling

MRI

PEDIATRIC

^1|2Indicates the priority used for review

Provide references using author date format

Marquand, A. F., Rezek, I., Buitelaar, J., & Beckmann, C. F. (2016). Understanding heterogeneity in clinical cohorts using normative models: Beyond case-control studies. Biological Psychiatry, 80, 552–561.
Di Biase, M. A., Tian, Y. E., Bethlehem, R. A. I., Seidlitz, J., Alexander-Bloch, A. F., Yeo, B. T. T., & Zalesky, A. (2023). Mapping human brain charts cross-sectionally and longitudinally. Proceedings of the National Academy of Sciences of the United States of America, 120(20), e2216798120.
Stasinopoulos, M. D., Rigby, R. A., Heller, G. Z., Voudouris, V., & De Bastiani, F. (2017). Flexible Regression and Smoothing: Using GAMLSS in R. CRC Press.
Gilmore, J. H., Knickmeyer, R. C., & Gao, W. (2018). Imaging structural and functional brain development in early childhood. Nature Reviews Neuroscience, 19, 123-137.