UPM Institutional Repository

Machine learning models to revealing the impacts of ibuprofen on Fistulifera pelliculosa and Mesolimbus marine algae


Citation

Bingke, Yang and Zhou, Yong and Lei, Wu and Amirian, Veghar and Russel, Mohammad and Zhang, Dayong and Yusof, Zetty Norhana Balia (2025) Machine learning models to revealing the impacts of ibuprofen on Fistulifera pelliculosa and Mesolimbus marine algae. Journal of Environmental Chemical Engineering, 13 (3). art. no. 116859. pp. 1-17. ISSN 2213-3437

Abstract

This study intelligently analyzes the effects of ibuprofen (IBU) on the metabolism of Fistulifera pelliculosa and Mesolimbus marine algae, through the integration of machine learning (ML) and conventional analytical methods. Malondialdehyde (MDA) and Superoxide dismutase (SOD) had the highest level (0.282 nmol/L and 2.461 U/ML, respectively) for Mesolimbus under exposure to 100 mg/L IBU. The lowest OD686, Chl-a, and fucoxanthin values were 0.221, 264.89, and 9.531 mg/L respectively, at 100 mg/L IBU. The ML-integrated approach revealed a smart graphical pattern of OD686, Chl-a, and sugar, protein, lipid & fucoxanthin metabolites, highlighting IBU's impact on algae. Moreover, K-means cluster analysis grading the 0-0.1 < 1-10 < 50-200 mg/L groups of IBU influence on algal growth activity. Support vector machine (SVM) confirm > 5.0 mg/L IBU influenced the growth biomass activity, the accuracy and recall are both 100 %, and the decision boundary determines the threshold 4.8 mg/L concentration of IBU affecting Mesolimbus, which is faster and more precise compared to conventional techniques. The principal component analysis (PCA) results indicate that IBU primarily affects fucoxanthin metabolic products in the algae to inhibit algal growth, reduce chlorophyll content, and exacerbate oxidative stress in the algae, leading to increased levels of MDA and SOD. The hierarchical cluster analysis (HCA) and K-means models further confirm that IBU greater than 5.0 mg/L have a significant impact on fucoxanthin in both Fistulifera pelliculosa & Mesolimbus algae, and the effect intensified with increasing IBU concentration. ML models mitigate the conventional analyses challenges to enhancing the IBU impact assessment for environmental monitoring decision making tools.


Download File

[img] Text
121274.pdf - Published Version
Restricted to Repository staff only
Available under License Creative Commons Attribution.

Download (12MB)

Additional Metadata

Item Type: Article
Divisions: Faculty of Biotechnology and Biomolecular Sciences
DOI Number: https://doi.org/10.1016/j.jece.2025.116859
Publisher: Elsevier
Keywords: Hierarchical cluster analysis; Ibuprofen; Machine learning; Marine algae; Principal component analysis; Support vector machine
Depositing User: Ms. Nuraida Ibrahim
Date Deposited: 30 Oct 2025 03:44
Last Modified: 30 Oct 2025 03:44
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.1016/j.jece.2025.116859
URI: http://psasir.upm.edu.my/id/eprint/121274
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item