Nondestructive detection of fish freshness during its preservation by combining electronic nose and electronic tongue techniques in conjunction with chemometric analysis

• DOI: 10.1039/c3ay41579a
• Publication Date: 2013-10-29T10:30:24Z
• Authors: Fangkai Han; Xingyi Huang; Ernest Teye; Feifei Gu; Haiyang Gu
• Keywords: Electronic Nose; Electronic tongue; Training set; Data set

A new method was developed to detect fish freshness nondestructively by combining electronic nose (E-nose) and tongue (E-tongue) in conjunction with chemometric methods. An E-nose nine metal oxide semiconductor gas sensors a commercial E-tongue were employed this research. Pseudosciaena crocea stored at 4 °C for different days used as experimental samples. Total viable counts (TVC) of the detected conventional method. data analyzed principal component analysis. Three-layer radial basis function neural network (RBF-NN) models established qualitative discrimination freshness. Performances RBF-NN numbers components (PCs) input compared. Experimental results revealed that best model acquired seven PCs an optimal performance 87.9% 80.0% training set prediction respectively. While, analysis five 86.3% 81.8% set. Another built combination E-tongue. The result shows rates improved 94.0% 93.9% support vector machine regression applied establish relationship between combined from TVC values quantitative determination. high correlation found merged parameter coefficients more than 0.91. proved that, single system enough classify samples on °C, while higher rate two sensors. could also be quantitatively evaluate In conclusion, appropriate methods can conveniently °C.