CHEMOMETRIC SMART APPROACHES USING ARTIFICIAL NEURAL NETWORKS AND CONTINUOUS WAVELET TRANSFORM FOR SIMULTANEOUS QUANTITATIVE ANALYSIS OF CIPROFLOXACIN-ORNIDAZOLE TABLETS

Abstract

New chemometric smart approaches, Artificial Neural Network (ANN) and Continuous Wavelet Transform (CWT), based on UV spectrophotometric data, were proposed for the simultaneous quantitative analysis of ciprofloxacin and ornidazole in tablets. Both methods enabled the study of the two-component mixtures containing these drugs without requiring a pre-separation process. The ANN calibration model was developed by establishing a relationship between the absorbance measurement matrix and the calibration set, which was constructed using a full factorial design methodology. To quantify ciprofloxacin and ornidazole, Symlets8 continuous wavelet transform (sym8-CWT) exhibited to be a suitable tool for transforming the UV spectra during the calibration and prediction stages. Both chemometric methods were applied within the linear working range of 3–24 μg/mL for ciprofloxacin (CIP) and 6–32 μg/mL for ornidazole (ORN). The validity of the proposed ANN and sym8-CWT approaches was confirmed through the analysis of independent test samples, as well as intra-day, inter-day, and standard addition experiments. The ANN method provided impressive recovery rates of 99.9% for CIP and 100.1% for ORN. Similarly, the sym8-CWT method achieved recovery rates of 98.5% for CIP and 101.5% for ORN. Both ANN and sym8-CWT approaches were successfully applied to the real sample analysis of CIP-ORN tablets, demonstrating precise and accurate results at a low cost and with minimal sample preparation.