Predictive Intelligence for Mpox Virus Control in the Democratic Republic of Congo: Opportunities, Equity, and Pathways Forward

Machine learning (ML) and predictive intelligence are increasingly recognized as transformative tools for epidemic preparedness, yet their translation into operational use in Mpox-endemic and resource-limited settings such as the Democratic Republic of Congo (DRC) remains limited. This narrative review examines emerging evidence on opportunities, challenges, and equity considerations for integrating predictive approaches into Mpox control. We highlight principles of spatiotemporal learning, hybrid and ensemble modeling, and geospatial risk mapping under conditions of sparse and heterogeneous data. Rather than benchmarking algorithms, the review emphasizes context-sensitive priorities, including covariate selection, interpretability, and alignment with One Health frameworks. Pathways forward include transparent evaluation, participatory model development, and investment in local analytic capacity. By situating predictive intelligence within DRC’s health system realities, we argue that these approaches hold promise for strengthening early warning, improving targeted interventions, and advancing equitable epidemic preparedness.