Free Access
Issue
Analusis
Volume 28, Number 9, November 2000
Endocrine disruptors
Page(s) 825 - 829
Section Original articles
DOI https://doi.org/10.1051/analusis:2000150
Analusis 28, 825-829 (2000)
DOI: 10.1051/analusis:2000150

Pyrolysis-mass spectrometry for rapid classification of oysters according to rearing area

M. Cardinal1, C. Viallon2, C. Thonat2 and J.-L. Berdagué2

1  Laboratoire Génie Alimentaire, IFREMER, rue de l'île d'Yeu, BP 21105, 44311 Nantes Cedex 3, France
2  Laboratoire Flaveur, Station de Recherches sur la Viande, INRA de Theix, 63122 St-Genès-Champanelle, France


(Received July 6, 2000; revised November 13, 2000; accepted November 14, 2000.)

Abstract
Current concern for the safety and traceability of food, as well as the desire of oyster farmers, for marketing reason, to emphasise the geographical origin of their production, requires new methods to make possible a real product identification. In this study, 181 oyster samples were analysed to determine their origin area. These samples were collected in nine French rearing areas at four different times of the year (spring, summer, and the beginning and end of autumn) and from four to eight sites in each area to provide a variability parameter. Analysis of fingerprints after Curie point pyrolysis-mass spectrometry, by an artificial neural network gave a mean classification rate of 89 %. Although the technique requires further improvements, it appears to be a useful discriminative tool for rapid identification of an oyster production area.


Key words: Oyster -- classification -- pyrolysis-mass spectrometry -- production area -- neural network.


© EDP Sciences, Wiley-VCH 2000

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.