Differentiation of two Canary DO red wines according to their metal content from inductively coupled plasma optical emission spectrometry and graphite furnace atomic absorption spectrometry by using Probabilistic Neural Networks

The metal content of 54 commercialized wines (30 samples from Tacoronte-Acentejo DO (class T) and 24 Valle de la Orotava DO (class O) wines) was performed by ICP-OES (Al, Ba, Cu, Fe, Mn, Sr, Zn, Ca, K, Na and Mg) and GF-AAS (Ni and Pb). Wine samples were processed by dry ashing followed by solution with 5% nitric acid. Metals were considered as suitable descriptors to differentiate between T and O classes. Supervised learning pattern recognition procedures were applied. Linear discriminant analysis (LDA) led to good results up to about 90% of correct classification. In order to improve the results, another kind of algorithms able to model non-linear separation between classes was considered: Probabilistic Neural Networks. Accordingly, excellent results were obtained, leading to sensitivities and specificities higher than 95% for the two classes.