Pharmaceuticals and personal care product modelling: Unleashing artificial intelligence and machine learning capabilities and impact on one health and sustainable development goals

Sci Total Environ. 2024 Dec 10:955:176999. doi: 10.1016/j.scitotenv.2024.176999. Epub 2024 Oct 19.

Abstract

The presence of pharmaceutical and personal care products (PPCPs) in the environment poses a significant threat to environmental resources, given their potential risks to ecosystems and human health, even in trace amounts. While mathematical modelling offers a comprehensive approach to understanding the fate and transport of PPCPs in the environment, such studies have garnered less attention compared to field and laboratory investigations. This review examines the current state of modelling PPCPs, focusing on their sources, fate and transport mechanisms, and interactions within the whole ecosystem. Emphasis is placed on critically evaluating and discussing the underlying principles, ongoing advancements, and applications of diverse multimedia models across geographically distinct regions. Furthermore, the review underscores the imperative of ensuring data quality, strategically planning monitoring initiatives, and leveraging cutting-edge modelling techniques in the quest for a more holistic understanding of PPCP dynamics. It also ventures into prospective developments, particularly the integration of Artificial Intelligence (AI) and Machine Learning (ML) methodologies, to enhance the precision and predictive capabilities of PPCP models. In addition, the broader implications of PPCP modelling on sustainability development goals (SDG) and the One Health approach are also discussed. GIS-based modelling offers a cost-effective approach for incorporating time-variable parameters, enabling a spatially explicit analysis of contaminant fate. Swin-Transformer model enhanced with Normalization Attention Modules demonstrated strong groundwater level estimation with an R2 of 82 %. Meanwhile, integrating Interferometric Synthetic Aperture Radar (InSAR) time-series with gravity recovery and climate experiment (GRACE) data has been pivotal for assessing water-mass changes in the Indo-Gangetic basin, enhancing PPCP fate and transport modelling accuracy, though ongoing refinement is necessary for a comprehensive understanding of PPCP dynamics. The review aims to establish a framework for the future development of a comprehensive PPCP modelling approach, aiding researchers and policymakers in effectively managing water resources impacted by increasing PPCP levels.

Keywords: AI and ML; GIS; Multimedia modelling; One health; PPCPs; SDG.

Publication types

  • Review

MeSH terms

  • Artificial Intelligence*
  • Cosmetics* / analysis
  • Environmental Monitoring / methods
  • Humans
  • Machine Learning*
  • Models, Theoretical
  • Pharmaceutical Preparations
  • Sustainable Development*

Substances

  • Pharmaceutical Preparations
  • Cosmetics