ARTIFICIAL INTELLIGENCE FOR CLIMATE-SENSITIVE DISEASE SURVEILLANCE IN LOW- AND MIDDLE-INCOME COUNTRIES: A SYSTEMATIC LITERATURE REVIEW

Abstract

Climate-sensitive diseases, including malaria, dengue and cholera, are increasingly becoming threats to the lower and middle-income countries (LMCs) where variability in climate and weak health systems increase their vulnerability. Increasing temperatures, altered rain patterns, and severe weather patterns will be redefining the patterns of transmission of vector and water crush diseases in regions such as Sub-Saharan Africa, South Asia, Latin America, and others. In this regard, artificial intelligence (AI) is becoming trendy because it can help rectify the failure in the early detection of a disease by conducting predictive models and integrating climatic-health data.

This systematic literature review discusses how AI methods are used in climate-sensitive disease surveillance in LMICs. On the basis of the PRISMA guidelines, we have screened articles within the period of 2010-2024 through different databases such as PubMed, Scopus, Web of Science, and IEEE Xplore by using predetermined keywords and inclusion/exclusion criteria. There were thirty-nine studies that satisfied inclusion threshold.

Results indicate that machine learning algorithms, especially Random Forest, Support Vector machines (SVM) and deep learning, are commonly used to predict outbreaks on the basis of climate factors containing temperature, rainfall and humidity. The most common diseases to be studied were malaria and dengue. Nevertheless, there are still major gaps as the multi-country models are limited, the level of integration between climate datasets is insufficient and ethical aspects have not been studied enough.

According to the review, AI has the potential to reinforce the early warning systems and inform climate-resilient public health strategies. The explainability, data equity, and enhanced cross-sectorial collaboration are to be the subjects of future research.