Intelligent machine learning enabled sensor for acyclovir using NiMnO3 flower-like electrocatalyst


Bux N., Hussain S., KÜÇÜKDENİZ T., Ali Soomro R., A. M. Mersal G., KARAKUŞ S., ...More

Materials Science and Engineering: B, vol.309, 2024 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 309
  • Publication Date: 2024
  • Doi Number: 10.1016/j.mseb.2024.117668
  • Journal Name: Materials Science and Engineering: B
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Chemical Abstracts Core, Chimica, Communication Abstracts, Compendex, INSPEC, Metadex, Civil Engineering Abstracts
  • Keywords: Antiviral drugs, Bimetallic oxides, Electrocatalyst, Emerging pollutant, Machine learning
  • Istanbul University Affiliated: No

Abstract

Hierarchical flower-like NiMnO3 was prepared using a hydrothermal route for oxidative detection of acyclovir (ACV), an antiviral pharmaceutical pollutant. The flower-like structure with an electroactive surface area (EASA) of 0.0952 cm2 enables non-revisable oxidation of ACV, with differential pulse voltammetry (DPV) and amperometry confirming its robust analytical capability in both high (15–75 µM) and low (0.1–1.0 µM) concentration ranges, respectively. Using amperometry, the sensor achieved an estimated limit of detection (LOD) of 1.59 nM (S/N=3) with selective oxidation of ACV and a sensitivity of 1.039 µA µM−1 cm2 in the presence of other common interferants. The adaptation of machine learning (ML) algorithms like random forest, XGBoost, linear regression, and ANN validated sensors’ performance and confirmed ANN's superiority in DPV signal interpretation. NiMnO3, as an electrocatalyst for ACV oxidation, validated by ANN modeling, highlights bimetallic oxides’ potential as a cost-effective, versatile platform for detecting pharmaceutical pollutants.