Research priorities for data science and artificial intelligence in global health: an international consensus exercise

Applications of data science and artificial intelligence (AI) in global health are expanding, yet research remains fragmented and often misaligned with the needs of low-income and middle-income countries (LMICs). To address this misalignment, we conducted a global research priority-setting exercise using the Child Health and Nutrition Research Initiative (CHNRI) method. 155 research ideas were scored by 51 experts based on feasibility, potential impact on disease burden, paradigm shift potential, implementation potential, and equity. Top-ranked priorities focused on epidemic preparedness, including AI-based outbreak prediction, improved diagnostics for infectious diseases, and early-warning systems. Other highly ranked topics included AI-assisted resource allocation, telemedicine, culturally adapted mobile health services, and chronic disease management tools. Experts from LMICs prioritised infectious disease control and diagnostic equity, whereas experts from high-income countries emphasised infrastructure and climate-related analytics. The resulting agenda provides a roadmap for aligning AI and data science research with global health priorities, particularly in LMICs.