Cover of Susana Carmona's Book , Neuromaternal with a "+" followed by Chan Zuckerberg Initiative Logo

The Chan Zuckerberg Initiative has awarded the WBHI AI Core with $1,000,000 in funding to develop an AI model of the maternal brain. By leveraging multimodal geometric deep learning, the AI Core will create 3D models (dubbed “atlases” or “templates”) of the maternal brain. These models adapt based on hormone levels, physiology, and demographics to identify patterns of brain evolution, uncover population-level differences, and predict postpartum conditions. Resources will be dedicated to ensure the the models are explainable and fair, accurately representing diverse populations. 

The base of this work is the WBHI’s Maternal Brain Project (MBP) dataset, a pioneering resource that maps neuroanatomical changes in first time mothers in the U.S, and Spain from preconception through the first two years postpartum. Using this multimodal data, the MBP provides the foundation for this AI effort. Together, these insights and advancements aim to revolutionize our understanding of maternal mental health.

This project is carried out by our AI Core, with leadership from Nina Miolane, Susana Carmona, Emily Jacobs, and Magdalena Martinez Garcia and in collaboration with the Chan Zuckerberg Initiative.

Picture Credit: Susana Carmona’s Book, Neuromaternal.