Burst of gyrification in the human brain after birth

Abstract

Gyrification, the intricate folding of the brain's cortex, begins mid-gestation and surges dramatically throughout the perinatal period. Yet, a critical factor has been largely overlooked in neurodevelopmental research: the profound impact of birth on brain structure. Leveraging the largest known perinatal MRI dataset-819 sessions spanning 21 to 45 postconceptional weeks-we reveal a burst in gyrification immediately following birth (~37 weeks post-conception), amounting to half the entire gyrification expansion occurring during the fetal period. Using state-of-the-art, homogenized imaging processing tools across varied acquisition protocols, and applying a regression discontinuity design approach that is novel to neuroimaging, we provide the first evidence of a sudden, birth-triggered shift in cortical development. Investigation of additional cortical features confirms that this effect is uniquely confined to gyrification. This finding sheds light onto the understanding of early brain development, suggesting that the neurobiological consequences of birth may hold significant behavioral and physiological relevance.

Fig. 1. Consistent segmentations across fetal and postnatal participants.

Examples of segmentations for three fetuses and two postnatal participants illustrating the consistency of the anatomical delineation despite large variations in age and developmental stage. eCSF represents external cerebrospinal fluid, cGM represents cortical gray matter, WM represents white matter, dGM represents deep gray matter.

Fig. 2. Sharp change at the time of birth across the perinatal trajectory of gyrification and its components.

Plot illustrating the regression discontinuity (effect size = 7.48, p = 6.44 × 10⁻¹⁴) of gyrification in (A), and the regression discontinuities of its components (surface area: effect size = 3.68, p = 2.26 × 10⁻⁴; and convex hull area: effect size = 1.26, p = 0.21) in (B) as a function of age at a cut-off of 37 postconceptional weeks (time of birth) (n = 819). In (A), red dots represent the fetal sample and cyan dots represent the postnatal sample, with a line indicating the time of birth. In (B), purple points represent the surface area trajectory, green points represent the convex hull trajectory, dots represent the fetal sample, triangles represent the postnatal sample, and the dotted line indicates the time of birth.

Fig. 3. The sharp change in gyrification at birth is illustrated by the relationship between effect size and p value.

Effect size as a function of log-transformed p value from the regression discontinuity analysis in each feature (a higher effect size indicates a larger discontinuity at our cut-off, i.e., at birth at 37 weeks). Gyrification is shown to have a strikingly high effect size compared to other features.

Fig. 4. Preterm subjects mapped onto the perinatal trajectory of gyrification.

We show that the neurodevelopment of gyrification as a function of age for preterm participants (n = 102) follows a continuous trajectory compared to a jump seen around birth in our typically developing population (n = 819).