AUTHOR=Ciceri Tommaso , Squarcina Letizia , Bertoldo Alessandra , Brambilla Paolo , Melzi Simone , Peruzzo Denis 

TITLE=Fetal gestational age prediction via shape descriptors of cortical development

JOURNAL=Frontiers in Pediatrics

VOLUME=Volume 12 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2024.1471080

DOI=10.3389/fped.2024.1471080

ISSN=2296-2360

ABSTRACT=Gyrification is the intricate process through which the mammalian cerebral cortex develops its characteristic pattern of sulci and gyri. This process commences around the 10th week of gestation in human fetuses and continues hierarchically even after birth. Monitoring gyrification provides valuable insights into brain development and identifies potential abnormalities at an early stage. This study analyzes the cortical structure using various shape descriptors to understand the gyrification process during pregnancy. We compare the morphometric properties encoded with the commonly used scalar point-wise curvature-based signatures (mean curvature H, Gaussian curvature K, shape index SI, curvedness C) to multidimensional point-wise shape signatures derived via techniques developed for computer graphics and rarely applied to the medical field. These multidimensional point-wise shape signatures include spectral geometry processing methods such as Heat Kernel Signature (HKS) and Wave Kernel Signature (WKS), as well as the intersection between histogram and signature techniques known as Signature of Histograms of OrienTations (SHOT). This work proposes a novel technique to extract a global descriptor from a given point-wise signature: GHKS, GWKS, and GSHOT. The extracted signatures are evaluated using Support Vector Regression (SVR)-based algorithms to predict the fetal gestational age (GA). Results show that GSHOT better encodes the GA to other global multidimensional point-wise shape signatures (GHKS, GWKS) and commonly used scalar point-wise curvature-based signatures (C, H, K, SI, FI). These results allow for new insights into neuroimaging for a better understanding of fetal brain development and, in particular, of the gyrification process.