The MNCH Project will apply novel approaches to data assimilation and advanced AI/ML-based methods to develop early warning systems that identify mothers and children at risk for poor health trajectories. We will leverage the existing large datasets from studies being conducted by our Hub partners and utilize novel data science approaches to develop predictive and prescriptive models for common maternal and childhood causes of morbidity and mortality in SSA.

• Using data from a longitudinal surveillance system i.e., Kilifi Perinatal and Maternal Research Project (KIPMAT) and an ongoing, large-scale cohort study i.e., PREgnancy Care Integrating translational Science, Everywhere (PRECISE), we will develop and validate AI/ML-based prediction models for poor pregnancy outcomes (e.g., placental disorders, low birthweight) among Kenyan women.
• We will also leverage the KIPMAT dataset to develop and validate AI/ML-based prediction models to determine the long-term risk of developing cardiovascular morbidity and mortality among mothers with a history of pregnancy-induced hypertension and/or gestational diabetes.
• Lastly, we will identify trajectories of long-term life outcomes (e.g., neurodevelopmental disorders, school performance, etc.) among Kenyan children using predictive models developed from the Kenya Medical Research Institute-Wellcome Trust Research Programme’s Early Childhood Development (ECD) dataset.

This project will lay the groundwork for our Hub platform to be scaled up and actualized through further collaborations with technological partners in SSA in the form of innovations (e.g., mobile phone applications) that will be able to bring individual-level maternal and child health risk prediction capability to the hands of various stakeholders (e.g., frontline clinical providers as well as public health officials).