Integrating Machine Learning, Deep Learning, and Geospatial Data Science for Public Health Surveillance and Health Inequality Assessment: A Scoping Review

Abstract

Combining machine learning (ML), deep learning (DL) with geospatial data science (GIS) holds potential for improving public health surveillance and the evaluation of health inequalities. The scoping review examines their use in disease prediction, spatial risk mapping, outbreak detection and health facility accessibility analysis. It explores geospatial data sources, analytic methods and the possibilities of using an integrated model to identify vulnerable social groups and geographical health inequalities. Some of the major challenges are data quality, algorithmic bias, privacy, model interpretability, and computational resources. The review identifies research priorities to build on robust, fair, and transparent analytical tools to support evidence-based public health decision-making and mitigate health inequalities.