Understanding Barriers to Healthcare Delivery through Modeling of Maternity Healthcare Delivery in Kenya

The challenge of improving maternal healthcare outcomes in Kenya requires an approach that comprehensively acquires and represents knowledge related to the public healthcare landscape, placing observed data and outcomes in context to determine why women are not receiving the care that is available to them. Without the ability to structure and connect various data, capabilities are limited in their ability to express meaningful links between the times, places, and context that may be key to understanding the processes that influence maternal health outcomes. To address these challenges, we propose to develop and demonstrate the feasibility of an integrated public health data lake and knowledge graph decision support system. The combination of a scalable storage repository that manages large streams of heterogeneous data and a metadata knowledge graph that enables structured direct and contextual data representation will provide a platform for generating key insights for understanding maternal healthcare population segmentation, and drive future insights that can inform recommendations to providers for improving maternal health outcomes. By combining the strengths and expertise of Orsalus, IST Research, and Milcord, we have the capability to build a proof of concept in a real-world operating environment and demonstrate the value of this approach.