A Nondestructive Detection Method for Mixed Veterinary Drugs in Pork Using Line-Scan Raman Chemical Imaging Technology

• Published in: Food analytical methods 2019-03, Vol.12 (3), p.658-667
• Main Authors: Wang, Wenxiu, Zhai, Chen, Peng, Yankun, Chao, Kuanglin
• Format: Article
• Language: English
• Subjects: Analytical Chemistry; Chemistry; Chemistry and Materials Science; Chemistry/Food Science; Chloramphenicol; Chloromycetin; Correlation coefficients; Drugs; Food Science; Image acquisition; Image processing; Microbiology; Nondestructive testing; Ofloxacin; Organic chemistry; Pixels; Pork; Quantitation; Quantitative analysis; Regression analysis; Regression models; Spatial distribution; Sulfadimidine; sulfamethazine; Technology; Veterinary drug residues; veterinary drugs; Veterinary medicine
• ISSN: 1936-9751; 1936-976X
• DOI: 10.1007/s12161-018-1397-5

This study reports a nondestructive detection method using Raman chemical imaging (RCI) technology for the simultaneous determination of multiple veterinary drugs, such as ofloxacin, chloramphenicol, and sulfadimidine, in pork. A line-scan Raman imaging system was employed to acquire images of pure veterinary drugs and pork samples containing single and mixed drugs. Raman characteristic peaks at 1623, 1353, and 1147 cm −1 were identified for ofloxacin, chloramphenicol, and sulfadimidine, respectively. An image processing method was proposed to calculate the “pixel-ratio” values in each feature image, and linear regression models for a single veterinary drug were then established between the “pixel-ratio” values and the actual concentrations. By applying the models to samples with mixed veterinary drugs, the concentrations of the three drugs were predicted with correlation coefficients of 0.978, 0.986, and 0.984. The satisfactory results indicated that along with the proposed “pixel-ratio” method, RCI technology enables the nondestructive quantitation and spatial distribution visualization of multiple veterinary drug residues.