ArticlePDF Available

Use of the electronic nose on products of Cinta Senese pigs

Taylor & Francis
Italian Journal of Animal Science
Authors:

Abstract

The use of a quartz microbalance based (QMB) electronic nose for feed traceability of fresh and cured fat of Cinta Senese pigs has been evaluated. Thirty-three pigs were fed different feeding during fattening: “three months chestnut” (3-CH), “1 month chestnut” (1-CH) “fed commercial feedstuff” (0-CH). Fresh fat and cured lard of each animal were analysed. Overall data set was analysed by factorial analysis to test if the instruments allowed a satisfactory pattern separation among groups. Afterwards, on the three factors generated by factorial analysis, a GLM procedure was applied to estimate effects such as: feeding type, operative temperature, day of analysis, order within day, layer of the subcutaneous fat. The results showed a clear separation according to feeding regimen in fresh fat only, especially between 1-CH and 0-CH, but also a strong effect of the other sources of variability. Concerning this, the date of analysis had a significant effect on each factor generated by factorial analysis that invalidated the discrimination obtained.
A preview of the PDF is not available
ResearchGate has not been able to resolve any citations for this publication.
Article
In this paper, we report a specific application of an electronic nose based on a semiconductor thin film sensor array for the discrimination of different hams. The sensing elements of the sensor array were undoped and doped (Cr, Pt) SnO2 thin films prepared by sputtering process. A system based on a static headspace sampling was used for the injection of the volatile compounds coming from the hams. Data analysis was performed by two pattern recognition methods: principal component analysis (PCA) and probabilistic neuronal network (PNN). The obtained results show that the electronic nose was able to identify the ham samples clearly.
Article
An electronic nose system to control the processing of dry-cured Iberian ham is presented. The sensors involved are tin oxide semiconductors thin films. They were prepared by RF sputtering. Some of the sensors were doped with metal catalysts as Pt and Pd, in order to improve the selectivity of the sensors. The multisensor with 16 semiconductor sensors, gave different responses from two types of dry-cured Iberian hams which differ in the feeding and curing time. The data has been analysed using the PCA (principal component analysis) and backpropagation and probabilistic neural networks. The analysis shows that different types of Iberian ham can be discriminated and identified successfully.