Computer-Aided Design of Eco-Friendly Imprinted Polymer Decorated Sensors Augmented by Self-Validated Ensemble Modeling Designs for the Quantitation of Drotaverine Hydrochloride in Dosage Form and Human Plasma.

The article focuses on the development and optimization of eco-friendly molecularly imprinted polymer (MIP)-decorated polyvinyl chloride (PVC) sensors for the selective and sensitive quantitation of drotaverine hydrochloride (DVN) in pharmaceutical dosage forms and human plasma. Utilizing a novel self-validated ensemble modeling (SVEM) machine learning approach combined with molecular dynamics and quantum mechanical (MD/QM) simulations, the study rationally designed MIP particles with magnetic Fe3O4 nanoparticle cores to enhance sensor performance. Four PVC membrane sensors were fabricated and optimized via different experimental designs, with the CCD and SVEM-Lasso models showing the best accuracy, sensitivity (Nernstian slopes ~59 mV/decade), and eco-friendliness as assessed by the AGREE metric. The sensors demonstrated good selectivity, rapid response, stability over six weeks, and were validated according to International Union of Pure and Applied Chemistry (IUPAC) guidelines, enabling reliable DVN detection in complex matrices such as spiked human plasma and pharmaceutical tablets.