Application of Artificial Intelligence in Detection and Remediation of Heavy Metals in Contaminated Water and Soils in Africa -A Systematic Review

Abstract

Heavy metal poisoning of water and soils remains a serious environmental and public health issue in Africa, fuelled by growing industrialisation, mining operations, urban growth, and insufficient waste management systems. Although artificial intelligence has been widely regarded as a transformative tool for global environmental monitoring, existing research rarely addresses Africa-specific constraints, rarely integrates detection and remediation, and offers limited guidance for policy implementation in data- and resource-constrained contexts. This study conducts a systematic review using a structured literature synthesis guided by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) screening principles to investigate artificial intelligence-based approaches for heavy metal detection, prediction, mapping, and remediation in African water and soil systems. The evaluation assesses the performance, data needs, and contextual applicability of machine learning, deep learning, adaptive neuro-fuzzy inference systems, Internet of Things-enabled sensing, and geospatial artificial intelligence in comparison to worldwide applications. The synthesis shows that remote sensing-artificial intelligence frameworks and Internet of Things-integrated systems enable scalable monitoring and early-warning capabilities, whereas artificial neural networks and adaptive neuro-fuzzy inference systems achieve high predictive accuracy (coefficient of determination frequently > 0.90) under limited data conditions. Despite showing potential, adoption is hampered by fragmented datasets, poor digital infrastructure, talent gaps, and insufficient integration with regulatory frameworks. To address these issues, the report suggests an Africa-focused artificial intelligence-environment policy framework to promote sustainable heavy metal management and increase environmental governance.