Submission ID 130668
| Issue/Objective | The use of artificial intelligence (AI) and machine learning (ML) continues to grow in global health, particularly in pandemic preparedness activities. While the approach offers significant promise, how these high-end technology-enabled surveillance systems translate into practice across diverse health system contexts remains poorly understood. This scoping review addresses this gap by examining how AI and ML are integrated into pandemic preparedness frameworks globally. |
|---|---|
| Methodology/Approach | Using the Arksey and O'Malley scoping review methodology and Braun and Clarke's thematic analysis, 38 peer-reviewed and grey literature pandemic preparedness frameworks published between January 2000, and June 2025 were systematically mapped. Findings were organized across four domains-Predict, Prevent, Detect, and Respond-based on the World Health Organization COVID-19 Strategic Preparedness and Response Plan. |
| Results | AI and ML applications were identified across multiple preparedness domains, including early warning and alert systems, real-time surveillance, automated detection algorithms, and ML-based resource allocation forecasting. While these tools enhanced timeliness and responsiveness, their integration was concentrated in global and high-resource national frameworks. Regional disparities were evident, with greater representation in the Western Pacific region and limited representation in regions such as the Eastern Mediterranean and the Americas. Pandemic preparedness frameworks incorporating AI and ML tend to assume infrastructural, data, and workforce prerequisites that are absent in many lower-resourced settings. Governance dimensions of AI use, including data interoperability and accountability mechanisms, were inconsistently addressed across the reviewed frameworks. |
| Discussion/Conclusion | AI-enabled surveillance holds genuine promise for advancing pandemic preparedness, but this review reveals that technological innovation in preparedness is currently outpacing its accessibility. The digital divide is particularly evident as regions with the highest infectious disease burden experience limited access. Realizing the full potential of AI-driven adaptive health security requires deliberate investment in digital infrastructure, context-sensitive governance frameworks, and the co-development of these tools with and for lower-resourced health systems. |
| Presenters and Affiliations | Dipika Shankar Bhattacharyya University of Waterloo Brent McCready-Branch University of Waterloo Labiqah Iftikhar University of Waterloo Thenugaa Rajeswaran University of Waterloo Jobaida Behtarin University of Waterloo Lakshna Ponrajah University of Waterloo Veronica Guglietti University of Waterloo Christina Mac University of Waterloo Md Akhtarul Islam University of Waterloo Zahid Ahmad Butt University of Waterloo |
