Data Challenge: AMR Predictive Analytics Engine (AMR-PAE)

Antimicrobial resistance (AMR) decreases the effectiveness of life-saving antibiotics and challenges patient care. To address this global problem, we propose developing a cutting-edge, AI-driven solution that leverages AMR surveillance datasets to deliver insights for AMR management and decision-making.
My goal is to integrate and harmonise data from multiple global AMR surveillance programs using machine learning to identify the most promising analytical strategy for actionable AMR insights.
Through this data-driven exploration, we hope to offer a solution with real-time, predictive insights that help clinicians make informed decisions, selecting the most effective treatments and avoiding antibiotics with high resistance rates. This approach will directly improve patient outcomes by reducing treatment failures and hospital stays. Moreover, by identifying areas of antibiotic overuse and misuse, the project aims to strengthen antimicrobial stewardship programmes, promoting responsible prescription practices.
For public health practitioners, the solution will generate interactive dashboards and geospatial visualisations that highlight regional and global resistance hotspots. This information will inform targeted interventions and resource allocation, contributing to stronger health systems and more effective AMR containment strategies.
This research will test and compare multiple analytical approaches to determine the most impactful solution for the Vivli AMR Data Challenge, ensuring that the final solution aligns with both technical feasibility and stakeholder needs.