Evaluation of extra-virgin olive oils shelf life using an electronic tongue: chemometric approach

Physicochemical quality parameters, olfactory and gustatory–retronasal positive sensations of extra-virgin olive oils vary during storage leading to a decrease in the overall quality. Olive oil quality decline may prevent the compliance of olive oil quality with labeling and significantly reduce shelf life, resulting in important economic losses and negatively condition the consumer confidence. The feasibility of applying an electronic tongue to assess olive oils’ usual commercial light storage conditions and storage time was evaluated and compared with the discrimination potential of physicochemical or positive olfactory/gustatory sensorial parameters. Linear discriminant models, based on subsets of 5–8 electronic tongue sensor signals, selected by the meta-heuristic simulated annealing variable selection algorithm, allowed the correct classification of olive oils according to the light exposition conditions and/or storage time (sensitivities and specificities for leave-one-out cross-validation: 82–96 %). The predictive performance of the E-tongue approach was further evaluated using an external independent dataset selected using the Kennard–Stone algorithm and, in general, better classification rates (sensitivities and specificities for external dataset: 67–100 %) were obtained compared to those achieved using physicochemical or sensorial data. So, the work carried out is a proof-of-principle that the proposed electrochemical device could be a practical and versatile tool for, in a single and fast electrochemical assay, successfully discriminate olive oils with different storage times and/or exposed to different light conditions.The authors acknowledge the financial support from the strategic funding of UID/BIO/04469/2013 unit, from Project POCI-01-0145-FEDER-006984—Associate Laboratory LSRELCM funded by FEDER funds through COMPETE2020—Programa Operacional Competitividade e Internacionalização (POCI)—and by national funds through FCT—Fundação para a Ciência e a Tecnologia and under the strategic funding of UID/BIO/04469/2013 unit. Nuno Rodrigues thanks FCT, POPH-QREN and FSE for the Ph.D. Grant (SFRH/BD/104038/2014).