Ultrasound (US) imaging is one of the first steps in a continuum of pregnancy care. During the fetal period, the brain undergoes dramatic structural changes, which are informative of healthy maturation. The resolution of modern US machines enables us to observe and measure brain structures in fetuses from as early as 14 weeks. Capitalizing on recent breakthroughs in machine learning, my group develops bespoke methods to automatically align brain images and track spatiotemporal patterns of intra-uterine brain development. In this talk, I will summarise our work on the design of data-driven techniques to build the first US-based atlas of the fetal brain. We envision that this atlas will serve as a population reference against which individuals can be compared, and hence enable detection of developmental deviations in routine clinical care.