Assessment of resting state structural-functional relationships in perisylvian region during the early weeks after birth

Abstract

This study investigates the structural-functional (S-F) relationships of the perisylvian region during the first weeks after birth in the resting state. Previous joint S-F studies of perisylvian development were mainly conducted on individual structural and functional metrics. By employing a weighted combination of metrics, joint S-F studies can enhance the understanding of perisylvian development in neonates, thereby providing valuable features for the prediction of neurodevelopmental disorders. To this end, we employed both structural and functional metrics derived from 16 perisylvian sub-regions (PSRs). Structural metrics included morphological and myelination measures, while functional metrics encompassed functional connectivity (FC) values between a designated seed PSR and the remaining PSRs. In addition, fractional amplitude of low-frequency fluctuations (fALFF) for all PSRs was computed and used as a distinct group of functional metrics. During statistical procedure, based on sparse canonical correlation analysis (CCA), the structural metrics were correlated with each respective group of functional metrics. Then, CCA was employed to delineate regional interdependencies and derive combined structural-functional features. The findings revealed that combined features outperform individual metrics for characterizing the normal development of the PSRs in term neonates. The myelination is prominently related to both fALFF and FCs, while morphological metrics of the PSRs have very limited contribution in combined structural features. Among the designated PSRs, the FCs of both the right insula and Heschls gyrus with other PSRs demonstrated a more robust correlation with the combined structural features.