Mapping infant brain functional regionalization with multiview functional cortical parcellation

1305 

Abstract Submission 

Tianli Tao¹, Jiawei Huang¹, Xinyi Cai¹, Feihong Liu¹, Rui Zhou¹, Zhuoyang Gu¹, Lianghu Guo¹, Lixuan Zhu¹, Feng Shi², Han Zhang¹

¹School of Biomedical Engineering, ShanghaiTech University, China, Shanghai, China, ²Shanghai United Imaging Intelligence Co., Ltd, Shanghai, China, Shanghai, China

Tianli Tao  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Jiawei Huang  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Xinyi Cai  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Feihong Liu  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Rui Zhou  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Zhuoyang Gu  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Lianghu Guo  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Lixuan Zhu  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

Feng Shi  
Shanghai United Imaging Intelligence Co., Ltd, Shanghai, China
Shanghai, China

Han Zhang  
School of Biomedical Engineering, ShanghaiTech University, China
Shanghai, China

The human brain demonstrates higher spatial and functional heterogeneity during the first two postnatal years than any other period of life (Johnson, 2000). Understanding the changes in brain function and structure has profound implications for comprehending early brain development. While existing research has provided substantial insights into the development of brain structure, our understanding of the development of cortical functions remains limited. Using advanced infant cortical functional parcellation, we investigate the relationship between functional regionalization and cognitive development. This work aims to address two questions: First, what is the functional hierarchy of brain functional regionalization? Second, what is the relationship between longitudinal functional regionalization and cognitive development?

A total of 467 scans collected from the Baby Connectome Project (BCP) (Howell, et al., 2019) were included in the current study. These scans comprised paired T1-weighted (T1w) MRI and resting-state fMRI data from 221 infants aged between 2 weeks and 24 months. Infant-based preprocessing procedures were applied to the T1w MRI and resting-state fMRI data. (Li, et al., 2013) We utilized our previously developed multiview functional parcellation with 9 brain subnetworks (Tao, et al., 2023) as the infant brain functional atlas. Consistent with the methodology outlined in the prior study, we calculated the multiview fMRI fingerprint and projected the data onto the individual cortical surface using Freesurfer (Fischl, 2012). Each subject's sphere has been registered to the fsaverage standard sphere, ensuring that corresponding vertices for each individual are aligned at the same location.
First, we employed a dendrogram analysis to explore the hierarchical relationships among the 9 subnetworks. Specifically, we utilized the multiview fMRI fingerprint to calculate the pairwise correlations among 9 subnetworks for each scan. Based on these correlations, we constructed a dendrogram to depict the hierarchical organization among the regions. Then, we utilized a linear mixed model to model the correlations between these subnetworks and five standardized Mullen scores (Mullen, 1995), namely Gross Motor (GM), Fine Motor (FM), Visual Reception (VR), Receptive Language (RL), and Expressive Language (EL).

Upon dendrogram analysis, we observed that the 9 subnetworks of the brain could be primarily categorized into three different clusters of functional development: 1) subnetwork 3 (Visual); 2) subnetworks 1, 7, 8 (Dorsal attention, Default mode, Parietal); and 3) subnetworks 2, 4, 5, 6, 9 (Limbic, Somatomotor, Superior temporal Frontal, Inferior temporal) (Fig. 1). This hierarchical organization suggests a prioritization in the functional development of different subnetworks, with a higher priority in the development of visual areas. This implies a more urgent need for visual capabilities during development in infancy.
As shown in Fig. 2, we found a significantly negative correlation between the subnetworks 1, 5, 7, 8 and receptive language scores, as well as between subnetworks 3, 9 and visual reception scores. No significant correlation was observed between any other Mullen scores and other subnetworks. Such results suggest that lower regional homogeneity, or higher functional diversity, maybe associated with enhanced information acquisition abilities (such as receptive language and vision) in infants. This finding may reflect the notion that functional differentiation within these specific brain subnetworks contributes to heightened perceptual and language capabilities.

Our study reveals a hierarchical organization in the brain functional development and its significant association with receptive language and vision abilities during early development. This suggests that the heterogeneity in functional development may contribute to elevated cognitive levels in infants.

Normal Brain Development: Fetus to Adolescence ¹

Normal Development ²

Cognition

Cortex

Cortical Layers

Development

MRI