An electronic tongue for fish freshness analysis using a thick-film array of electrodes

Author(s): Barat, J.M. | Escriche, I. | GarciaBreijo, E. | Gil, L. | MartinezManez, R. | Soto, J. |

Year: 2008

Citation: MICROCHIMICA ACTA Volume: 163 Issue: 1-2 Pages: 121-129

Abstract: We report on the development of an electronic tongue as potential tool for fish freshness determination. The studies were carried out following the evolution with time on fillet of cultured sea bream (Sparus Auratus). The electronic tongue consists of 16 potentiometric electrodes of the type metal, metal oxide and insoluble metal salts. Graphite was also used as electrode. Fish freshness indicators such as texture, pH, colour, microbial analysis, total volatile basic nitrogen (TVB-N) and biogenic amines were also determined versus time. The effectivity of the electronic tongue in the assessment of the evolution with time of fish fillets was evaluated. First the electronic tongue was used to classify samples according to time. Perceptron and adaptive resonance theory were used to discriminate between the data from different days. In a second step, the electronic tongue was used to predict the results obtained from chemical and biochemical analyses by building quantitative partial least square (PLS) models. A remarkable correlation was found between the electronic tongue formed by the 16 simple electrodes and parameters such as total biogenic amines, pH, TVB-N and microbial analysis with correlation coefficients larger than 0.98.

Topics: Machine Learning, Applications: Biological Classification, Models: Fuzzy ARTMAP,

PDF download

Cross References

  1. Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps
    A new neural network architecture is introduced for incremental supervised learning of recognition categories and multidimensional maps in response to arbitrary sequences of analog or binary input vectors, which may ... Article Details