Development of chemometric models using infrared spectroscopy (MID-FTIR) for detection of sulfathiazole and oxytetracycline residues in honey

Abstract

Chemometric models using mid-infrared (MID-FTIR) spectroscopy were developed for detection and quantification of oxytetracycline and sulfathiazole contamination in honey samples. Low standard error of calibration (SEC) and prediction (SEP) values were achieved using a partial least squares algorithm (SEC= 1.02 and SEP=1.39 for oxytetracycline and SEC=1.24 and SEP=1.79 for sulfathiazole). Chemometric model-predicted concentrations of antibiotics were compared with ELISA results with coefficient of determination R2=0.8577 for oxytetracycline and R2=0.8216 for sulfathiazole. Classification of antibiotic contaminated honey samples and uncontaminated samples was carried out using Soft Independent Modeling Class Analogy analysis with a 100% correct classification rate with interclass distances in the range of 6.93-13.3. MID-FTIR chemometric models developed for detection and quantification of oxitetracycline and sulfathiazole in honey samples have been demonstrated.