Abstract

Key points for the authentication of foodstuffs are the verification of the stated labelling, as well as checking its nature and verifying the absence of fraud. Portable spatially offset Raman spectroscopy (SORS) allows the rapid and straightforward acquisition of Raman spectra of packaged foods without the package opening and can therefore be considered as a non-invasive analytical technique par excellence since it enables in-situ analytical controls even directly on the grocery and supermarket shelves. Cheese is a dairy product available on the market which can be classified depending mainly on two characteristics features: (i) the animal origin of the milk from which it is produced and (ii) the manufacturing processes. This study aims to authenticate plastic-packaged sliced cheeses according to the origin of the milk stated by animal species (cow, sheep and/or goat). The NEAR similarity index was previously applied to verify that cheeses manufactured from the same animal species milk have a high similarity (NEAR >0.8), higher than those found among cheeses produced from milk of different animal species. From the adquired spectra, the one input-class SIMCA approach was used to develop multivariate classification models in order to distinguish the origin of the milk by animal species. The performance parameters of the developed classification models showed values close to 0.9 for sensitivity and precision among others. The PLSR method was used to develop quantification models capable of predicting the total protein and fat contents (major nutrients) in cheese samples having errors of less than 2%. These advances open the door to the development of new rapid and non-invasive analytical methods for in-situ packaged food authentication, demonstrating the potential of SORS for food authentication supported by proper chemometrics.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call