Bioinformatics and machine learning-based simultaneous spectrophotometric identification of Montelukast sodium and Desloratadine in a commercial film-coated tablet

DEMİRKAYA MİLOĞLU F., BAYRAK B., Kucuk H. B., ÖZBEK İ. Y., ÇETİN M.

JOURNAL OF ANALYTICAL SCIENCE AND TECHNOLOGY, cilt.16, sa.1, 2025 (SCI-Expanded, Scopus)

Özet

Desloratadine (DL) and montelukast sodium (MTS) are combined in a pharmaceutical preparation for the treatment of allergic rhinitis and asthma. DL is a competitive H1-receptor antagonist and helps the management of allergic reactions and relieves allergy symptoms, while MTS is a leukotriene receptor antagonist that inhibits the effects of inflammatory mediators. This study introduces a machine learning-assisted UV-Vis spectrophotometric method for the simultaneous quantification of DL and MTS in a commercial film-coated tablet, addressing limitations due to spectral overlap.Five machine learning regression models (Support Vector Regression, Ridge Regression, Lasso Regression, Elastic Net, and Linear Regression) were evaluated. Ridge Regression (lambda = 0.1) was selected for its balance of accuracy, computational efficiency, and robustness. The method was applied using Aircomb (R) film-coated tablets, containing 5 mg DL and 10.4 mg MTS, ensuring high precision.The developed method demonstrated high recovery rates (99.25% for DL and 101.0% for MTS with minimal relative standard deviation (<= 1.59%). Sustainability assessments using Analytical GREEnness Metric (AGREE) and Complex Green Analytical Procedure Index (ComplexGAPI) confirmed its alignment with green analytical chemistry principles. Ridge Regression (lambda = 0.1) provided accurate and reproducible results, making it suitable for routine pharmaceutical analysis.This study highlights the potential of machine learning-assisted UV-Vis spectrophotometry as a cost-effective and environmentally friendly alternative for pharmaceutical quality control. The method minimizes solvent consumption while ensuring analytical precision. Future research may explore non-linear models, such as artificial neural networks, to enhance predictive performance and broaden its applicability.