EUVIMEDEUVIMED entdecken
Übersichtsarbeit

Synergizing New Approach Methodologies and Artificial Intelligence: Charting the Paradigm Shift Toward Next-Generation Chemical Risk Assessment

Zidan Zheng, Yuyan Xu, Lixiang Xue, Xiao Chen, Yan Zhang, Jiayi Qian, Ziyi Huang, Hao Liu, Zhilang Li, Qibing Zeng

Environmental Pollution · 2027-01 · 412 · 129513

DOI: 10.1016/j.envpol.2026.129513 ↗

Kurz gefasst

Übersichtsarbeit in der Zeitschrift Environmental Pollution, 2027-01. Verfasst von Zidan Zheng und 9 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Trauma and Chemical Poisoning, National Institute for Occupational and Poison Control, Chinese Center for Disease Control and…. Die Arbeit nennt 111 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Beijing Municipal Health Commission.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The escalating global chemical inventory, coupled with the ethical and scientific inadequacies of traditional animal testing, has made a change in approach unavoidable in chemical risk assessment. This review examines how the combined use of New Approach Methodologies (NAMs) and Artificial Intelligence (AI) is reshaping this field. NAMs-including micro physiological systems, high-content screening, multi-omics, and in silico tools-generate high-dimensional and mechanism-anchored data. Concurrently, AI, spanning advanced machine learning and generative models, serves as the essential integrative engine that converts this data into actionable toxicological insight. We present a framework delineating how these two streams of technology interact to operationalize Next-Generation Risk Assessment (NGRA) across its entire pipeline-from intelligent hazard identification and refined exposure prediction to dynamic, population-aware risk characterization. The evidence indicates a shift from empirical, observation-based toxicology toward a predictive, mechanism-centric, and personalized science. By mapping specific AI techniques to core NAMs data streams and NGRA decision points across the hazard, exposure, and risk characterization continuum, this review provides a coherent roadmap for regulators, researchers, and industry stakeholders. We also consider environmental applications, including complex mixtures, persistent organic pollutants (POPs), per- and polyfluoroalkyl substances (PFAS), and emerging contaminants, while noting that the supporting evidence is more limited than for human health endpoints. Finally, we critically examine remaining challenges and chart forward-looking paths, including the development of chemical "digital twins," causal AI models, and real-time risk warning systems. We argue that progress depends on integrating NAMs and AI within a more efficient, ethical, and protective global chemical safety framework. However, barriers remain, including limited reproducibility across chemical domains, inadequate external validation, the hallucination risk of large language models in regulatory contexts, and the gap between academic development and regulatory acceptance. Open data infrastructure, standardized benchmarking, and harmonized validation frameworks will be essential to realize the full potential of this approach.

Schlagwörter

Stichwörter der Autor:innen

Artificial IntelligenceChemicalNew Approach MethodologiesNext-Generation Risk Assessment

Autor:innen und Institutionen

  • State Key Laboratory of Trauma and Chemical Poisoning, National Institute for Occupational and Poison Control, Chinese Center for Disease Control and Prevention, Beijing, 100050, China.
  • National Institute for Environmental Health, Chinese Center for Disease Control and Prevention, Beijing, 100021, China.
  • Center of Basic Medical Research, Institute of Medical Innovation and Research, Peking University Third Hospital, Beijing, 100191, China.
  • State Key Laboratory of Trauma and Chemical Poisoning, National Institute for Occupational and Poison Control, Chinese Center for Disease Control and Prevention, Beijing, 100050, China. Electronic address: zengqb@niohp.…

Förderung

  • National Natural Science Foundation of China
  • Beijing Municipal Health Commission

Zitieren

APA

Zheng, Z., Xu, Y., Xue, L., Chen, X., Zhang, Y., Qian, J., Huang, Z., Liu, H., Li, Z., & Zeng, Q. (2027). Synergizing New Approach Methodologies and Artificial Intelligence: Charting the Paradigm Shift Toward Next-Generation Chemical Risk Assessment. Environmental Pollution, 412, 129513. https://doi.org/10.1016/j.envpol.2026.129513

Vancouver

Zheng Z, Xu Y, Xue L, Chen X, Zhang Y, Qian J, et al. Synergizing New Approach Methodologies and Artificial Intelligence: Charting the Paradigm Shift Toward Next-Generation Chemical Risk Assessment. Environmental Pollution. 2027;412:129513. doi:10.1016/j.envpol.2026.129513

BibTeX
@article{Zheng2027,
  author = {Zheng, Zidan and Xu, Yuyan and Xue, Lixiang and Chen, Xiao and Zhang, Yan and Qian, Jiayi and Huang, Ziyi and Liu, Hao and Li, Zhilang and Zeng, Qibing},
  title = {Synergizing New Approach Methodologies and Artificial Intelligence: Charting the Paradigm Shift Toward Next-Generation Chemical Risk Assessment},
  journal = {Environmental Pollution},
  year = {2027},
  volume = {412},
  pages = {129513},
  doi = {10.1016/j.envpol.2026.129513}
}
Kostenlos starten →