Antibiotic mycelial residues (AMRs) added to animal feeds easily lead to drug resistance that affects human health and environment. However, there is a lack of effective detection methods, especially a fast and convenient detection technology, to distinguish AMRs from other components in animal feeds. To develop effective detection methods, two types of global Mahalanobis distance (GH) algorithms based on near-infrared microscopy (NIRM) imaging are proposed. The aim of this study is to investigate the feasibility of using NIRM imaging to identify AMRs in soybean meals. We prepared 15 mixed samples containing 5% AMRs using three types of soybean meals and four types of AMRs. The GH algorithm was used to identify non-soybean meals among the mixed samples. The hierarchical cluster analysis was employed to verify the recognition accuracy. The results indicate that use of the GH algorithm could identify soybean meals with AMR at a level as low as 5%.
Keywords: Antibiotic mycelial residue; global Mahalanobis distance; near-infrared microscopy; soybean meal.