Research Article

Artificial Intelligence as an Integrative Inference Layer in Syndemic Disease Surveillance

Abstract

The growing global burden of metabolic diseases alongside recurrent infectious disease outbreaks has exposed fundamental limitations in disease surveillance systems that remain siloed, reactive, and disease specific, despite increasing evidence of syndemic interactions driven by shared biological, social, and structural determinants. This review examines the role of artificial intelligence in enabling integrated metabolic and infectious disease surveillance, critically assessing current methods, identifying sources of bias and ethical risk, and articulating research and policy priorities for equitable and action oriented surveillance. We conducted a narrative scoping hybrid review of peer reviewed literature across public health, biomedical, and computational domains and analyzed studies using a conceptual framework that considered surveillance function, methodological approach, and equity and governance implications. Artificial intelligence methods including machine learning, deep learning, and network based models demonstrate substantial potential for multimodal data integration, early warning, and precision public health, but their real world impact is constrained by structural data inequities, algorithmic bias, limited interpretability, and weak integration into public health decision making. Performance metrics alone are insufficient to evaluate surveillance effectiveness, particularly with respect to equity, trust, and policy relevance. We conclude that artificial intelligence can enhance disease surveillance only if reconceptualized as public health infrastructure rather than isolated technological innovation, requiring equity centered design, interpretable and causal methods, robust governance, and interdisciplinary collaboration to avoid reinforcing existing disparities and to strengthen global health preparedness.