Artificial intelligence-driven approaches for detection, fate, transport, and remediation of microplastic pollution in aquatic systems

Microplastic pollution in aquatic systems is a global, pervasive and significant aquatic environmental problem with serious ecological and human health consequences. Traditional approaches to monitoring, analyzing, and modeling microplastics are labor-intensive, difficult to scale up and have limited predictive power. However, several large-scale studies have reported the occurrence, distribution and impact of microplastics. In the last few years, artificial intelligence, machine learning and deep learning have shown enormous potential to reshape how we monitor and manage the environment through data and automation. This narrative review critically synthesizes recent advances in artificial intelligence-assisted approaches for detection, characterization, environmental monitoring, fate and transport modeling, ecological risk assessment, and remediation of microplastic pollution in aquatic systems, with an emphasis on their current capabilities, limitations, and future research needs. Artificial intelligence-based image analysis and spectroscopic data processing have demonstrated improved accuracy and efficiency for automated microplastic identification under laboratory and controlled research conditions, although broader field validation remains limited. The use of artificial intelligence in evaluating exposure scenarios and potential impacts on biota and human health has emerged. Artificial intelligence-assisted optimization has shown promise for improving treatment-process performance in experimental studies; however, validation under full-scale operational conditions remains limited. However, several factors have hindered large-scale adoption, including data availability, model generalisability, standardization and interpretability. The study objectives were to synthesize current advances, highlight limitations, and suggest future research directions, such as the development of integrated artificial intelligence frameworks, digital twin technologies, and autonomous monitoring systems. By linking data science and environmental engineering, artificial intelligence-led strategies offer a revolutionary approach to more efficiently and sustainably managing microplastic pollution in aquatic systems. Unlike previous reviews that primarily address individual aspects of microplastic pollution or artificial intelligence applications, this review provides an integrated critical assessment of artificial intelligence -assisted detection, environmental monitoring, fate and transport modeling, ecological risk assessment, and remediation, while identifying key research gaps related to data standardization, model generalizability, field-scale validation, and interdisciplinary integration.

کلیدواژه ها:

Artificial intelligence (AI) ، Artificial intelligence-based detection ، Artificial intelligence-based remediation ، deep learning (DL) ، Environmental Monitoring ، Microplastics