Abstract: There is rapid development of the cerebral cortex that takes place during the perinatal period. In order to characterize this complex process spatio-temporal atlases are needed. At present, however, there is limited availability of cortical surface atlases for the infant brain. The development of robust tools applicable to this population has been significantly lagging compared to those introduced for adults given the difficulty of obtaining data from non-compliant neonates and toddlers and the rapid change in contrast and geometry displayed during development in infants. With the advent of large longitudinal brain imaging studies, such as the UNC/UMN Baby Connectome Project (BCP) and HEALthy Brain and Child Development (HBCD) Study, novel algorithmic solutions are needed to efficiently process these data sets. During the proposed project, we intend to develop a FreeSurfer-compatible pipeline that will establish a consistent and unbiased representation of the perinatal cortex over time and a set of tools that, when used with standard clinical MRI acquisitions, will enable the computation of accurate and robust spherical representations of the developing brain along with detailed macrostructural annotations. Such surfaces then will be used to characterize healthy brain development over the first ten years of life. For our clinical application, we will focus on the effects of early life adversity on brain development as well as on disentangling contradictory findings about brain abnormalities in very preterm infants, such as gyrification complexity in their anterior and posterior temporal lobes. This work will be performed at the MGH/Harvard/MIT Martinos Center for Biomedical Imaging relying on datasets collected by local collaborators, NIH-funded publicly available large scale data sets as well as a clinical cohort assembled by colleagues at the Washington University. Such an effort will allow us to take advantage of cutting-edge neonatal imaging and computational algorithm development expertise in an attempt to deliver computational tools robust and accurate enough for future clinical studies