Rapid Authenticity Testing of Alprazolam and Oxycodone Tablets Using Cloud-Based Portable Near-Infrared Technology

Abstract

This study evaluated a low-cost hand-held near-infrared (NIR) spectrometer for rapid, nondestructive authenticity screening of seized tablets. Spectra were acquired from 225 tablets across four alprazolam and two oxycodone finished dosage forms, comprising 57 authentic or consistent with authentic (CWA) tablets and 168 not consistent with authentic (NCWA) tablets. The spectra were preprocessed using second-derivative Savitzky-Golay and Standard Normal Variate (SNV), then analyzed using Euclidean distance metrics, and additionally, supervised partial least-squares-discriminant analysis (PLS-DA) and one-class SIMCA models. Authentic and CWA tablets exhibited consistently low spectral intravariability, whereas NCWA tablets showed markedly higher intervariability, enabling complete separation for all six imprints. Under tablet-level leave-one-tablet-out cross-validation, conservative distance thresholds gave 100% correct classification. Taking the falsified tablet as the positive class, PLS-DA achieved 99.4% sensitivity and 100% specificity, while one-class SIMCA achieved 100% sensitivity and 98.2% specificity. The method delivers a definitive CWA versus NCWA decision in under 1 min without sample preparation, offering law enforcement, customs, forensic laboratories, and harm-reduction services a first-line tool to triage high-risk falsified opioids and benzodiazepines.