Predictive Modelling for stillbirths and neonatal deaths in Sub-Saharan Africa

This project aims to address the high rates of stillbirths and neonatal deaths in Sub-Saharan Africa (SSA) by leveraging predictive modelling, including classical methods, machine learning (ML), and Artificial Intelligence (AI). SSA accounts for 47% global stillbirths and 46% neonatal deaths. Many SSA countries will unlikely meet their Sustainable Development Goals and Every Newborn Action Plan targets by 2030 due to poor quality of care, inadequate preventive measures, and data gaps. Current approaches are insufficient, but data science methodologies could mitigate these adverse outcomes. Using datasets from health-facilities and population studies across 15 African countries, I will develop robust predictive models for stillbirths and neonatal deaths. Advances in ML and AI are timely and can significantly improve predictions accuracy. Prediction models can enhance precision public health and resource efficiency, informing policy, planning, and clinical practice, ultimately reducing preventable stillbirths and neonatal deaths. Collaborations with academics, healthcare providers, and policymakers will ensure findings translate into actionable strategies. Outputs will include peer-reviewed publications, open-access datasets, and analytical tools, providing a framework for future research by integrating modern AI and ML techniques in SSA. Keywords: Stillbirths, Neonatal deaths, Predictive modelling, Machine learning, Artificial intelligence, Data science, Africa, Maternity registries, Demographic and Health Survey