Validation européenne de la procédure française de détermination par tomographie à rayons X de la nouvelle teneur en muscle des porcs

Gérard Daumas et al., 52e Journées de la Recherche Porcine (FRA), 4 et 5 février 2020, p. 55-56, poster

Poster.

Un nouveau règlement de l'UE sur la classification des carcasses s'applique à compter de juillet 2018. La nouvelle référence pour calibrer les méthodes de classification des porcs est un pourcentage de viande maigre basé sur la dissection manuelle totale (TMCdis) d'une demi-carcasse. La dissection manuelle peut être remplacée par une procédure de tomodensitométrie sans biais (RX). Daumas et Monziols (2018) ont calibré une méthode de tomodensitométrie simple, précise et robuste afin de la rendre utilisable au niveau international au moins en Europe. L’objectif de l’étude est de valider cette procédure de tomodensitométrie par un essai d'autorisation des méthodes de classement en Lituanie.

ENG

European validation of the French X-ray tomography procedure to determine new muscle content of pigs

Poster.

Since July 2018, the new benchmark for calibrating pig classification methods is muscle content based on total dissection of a halfcarcass (LMPdis). Manual dissection can be replaced by an unbiased X-ray tomography (CT) procedure. The purpose of this study was to validate the French procedure in Lithuania to make it usable at least in Europe. The CT procedure was initially calibrated with 29 French pork half-carcasses measured in 2008. The same CT and dissection procedures were applied to 12 Lithuanian half-carcasses sampled in 2018 as part of the approval of new classification methods of pork carcasses. A regression model was fitted between LMPdis and the CT LMP (LMPct). The intercept was not significant. For the calibration dataset, the slope of the regression was estimated at 0.965 and its 95% confidence interval (95CI) at [0.960, 0.970]. For the validation dataset, the slope of the regression was estimated at 0.967 and its 95CI at [0.954, 0.980]. The 95CI of the validation sample included the 95CI of the calibration sample, thus showing no statistical difference between the two slopes. The CT procedure developed by IFIP in France has been validated successfully in Lithuania for the approval of new national methods for classifying pork carcasses. From now on, this unbiased procedure should be able to be used in any EU Member State without further national dissection. In addition, this procedure can be used effectively for zootechnical trials without the risk of within-treatment bias.