Growth trajectories and cluster features of the human fetal brain estimated by signal intensity from sBTFE sequence in utero MRI

Abstract

Objective To evaluate the growth trajectories and cluster features of the human fetal brain tissues using signal intensity in utero magnetic resonance imaging. Methods In 136 fetal brain in utero, the signal intensity was measured in 20 region of interests (ROIs) including the bilateral thalamus, pons, genu, body and splenium of corpus callosum, the bilateral frontal, occipital, temporal pole, precentral and postcentral region, cerebellum and the bilateral posterior part of lateral ventricles as a reference. Calculating the signal intensity ratio (SIR) of these ROIs, and then, clustering analysis is used to explore the cluster features of SIR. Furthermore, five representative linear and nonlinear fitting models are used to estimate the relationship between SIR in each region and each cluster and gestational weeks. Results Over 22-38 gestational weeks, the SIR showed linear decrease in the most of (13 of 18) investigated brain regions, including the pons, bilateral occipital, precentral, postcentral, temporal, cerebellum, and thalamus. By using clustering analysis, the above brain regions could be classified into 6 clusters, and in 4 of 6 clusters, there was similar significant linear growth strategy with that in each of 13 of 18 regions. Conclusion The applications of a clinical available superfast sequence with sensitivity encoding with balanced turbo field echo sequence and signal intensity-based analysis may to some extent reflect the myelination process and growth strategies of fetal brain following heterogeneous and heterochronicity trajectories. It may be helpful for us to understand the normal fetal brain development pattern through in vivo imaging method prior to birth.