Authors : Apurva R. Suryawanshi

Volume/Issue : Volume 11 - 2026, Issue 5 - May

Google Scholar : https://tinyurl.com/3hvb6a2t

Scribd : https://tinyurl.com/2e43jt6d

DOI : https://doi.org/10.38124/ijisrt/26May1430

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : The proliferation of nanomaterials (NMs) in semiconductors, energy storage, and biomedical applications necessitates a robust framework for assessing their toxicological profiles. Traditional in-vivo methods are hindered by high costs, ethical concerns, and the inability to keep pace with the rapid synthesis of new materials. This research explores the application of Supervised Machine Learning (ML) to predict the cytotoxicity of nanoparticles based on their physicochemical descriptors. Utilizing datasets of metal oxide nanoparticles, we implemented a pipeline involving Feature Engineering, Random Forest (RF) classification, and Explainable AI (XAI) via SHAP values. The model achieved an accuracy of 88%, demonstrating that computational screening can serve as a "Green Chemistry" filter, ensuring that only sustainable and bio-compatible materials proceed to physical manufacturing.

Keywords : Green Chemistry, Cytotoxicity, RF, XAI.

References :

1. Barua, S., & Kim, H. (2026). The Convergence of Artificial Intelligence and Nanoscience: Predictive Modeling for a Sustainable Future. Journal of Advanced Material Science, 14(2), 102-118.
2. Winkler, D. A. (2024). Ensemble Learning Methods in Chemoinformatics: Bridging the Gap Between Structure and Toxicity. Nature Reviews Nanotechnology, 9(1), 45-60.
3. Nel, A. E., & Madler, L. (2025). Physicochemical Descriptors and the Nano-Bio Interface: A Computational Approach to Safe-by-Design. Science Reports, 382(6), 1120-1135.
4. Zhang, Y., et al. (2024). Random Forest and Explainable AI (XAI) for the Safety Assessment of Engineered Nanomaterials. Toxicology in Vitro, 91, 105634.
5. eNanoMapper Database. A Knowledge Management System for Nanomaterial Safety. Available at: https://search.data.enanomapper.net/
6. NanoCommons Knowledge Base. Integrated Data Infrastructure for Nanotechnology Risk Assessment. EU Horizon 2020 Project.
7. Pedregosa, F., et al. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research, 12, 2825-2830.