A Precision Public Health Approach to Improving Health Service Delivery for Maternal, Newborn, and Child Health in Rwanda: The Application of Predictive Machine Learning Algorithms and Innovative Data Visualization Techniques to Real-World Data

This project aims to improve the utilization of maternal, newborn, and child health (MNCH) data collected from a national program called RapidSMS in Rwanda. Since 2009, this program has employed community health workers (CHWs) across the country to collect health data on a mobile phone, sent via SMS to a central database. RapidSMS collects a large amount of data on MNCH daily, but these data are not optimally being utilized to inform health service delivery. An automated data cleaning process, application of machine learning algorithms, and the visualization of model outputs and key MNCH indicators are needed. MTEK Sciences (a data analytic group in Vancouver) and the Rwanda Biomedical Centre are proposing to collaborate on this project that will optimize in-country resource allocation for MNCH services, more responsive to local MNCH needs. Our project will be the first attempt at precision public health in MNCH, targeting communities at the micro-geographic level (5 by 5 km regions).