Article ID Journal Published Year Pages File Type
6397969 Food Research International 2014 7 Pages PDF
Abstract
A BioElectronic Tongue (BioET) based on a sensor array comprising 4 voltammetric enzyme-modified (bio)sensors plus pattern recognition and multivariate calibration data processing tools was applied towards the analysis of rosé cava wines. A total of 20 different samples from different producers were analysed using cyclic voltammetry without any sample pretreatment. Obtained responses were preprocessed employing the windowed slicing integral method in order to compress and extract significant features from the recorded data. Extracted coefficients were then evaluated by means of Principal Component Analysis to visualize some initial patterns, while quantification of different phenolic indexes was achieved by an Artificial Neural Network (ANN) model. In this manner, correlations were attempted between (bio)sensors responses and three different classical indexes related to total phenolic content (i.e. I280, I320 and Folin-Ciocalteu index) plus two other indexes related to other specific phenolic features (i.e. total tannins and anthocyanins content).
Related Topics
Life Sciences Agricultural and Biological Sciences Food Science
Authors
, , , ,