Explainable Machine Learning Models for AMR Prediction in Africa

Antimicrobial resistance (AMR) is one of the greatest threats to global health, with Africa facing unique challenges due to limited healthcare infrastructure, unregulated antibiotic use, and a high burden of infectious diseases. Common pathogens are increasingly resistant to antibiotics, making treatment options less effective and leading to higher mortality rates. There is a critical need for data-driven solutions to combat AMR, tailored specifically to African contexts.
This project will develop explainable machine learning (ML) models to predict AMR patterns in African countries. These models will not only predict resistance trends but will also provide insights into the key drivers of resistance using explainability techniques, enabling targeted interventions.
The explainable ML models will predict the likelihood of antibiotic resistance in pathogens, helping healthcare providers choose the most effective treatments for their patients.
By identifying the key factors driving resistance, the models will support the design of more effective stewardship programs.