Lean meat prediction with HGP, CGM and CSB-Image-Meater, with prediction accuracy evaluated for different proportions of gilts, boars and castrated boars in the pig population.

Prediction equations for the percentage lean meat in pig carcasses in The Netherlands were derived for the Hennessy Grading Probe 7, Capteur Gras/Maigre--Sydel and CSB-Image-Meater. Because castrated males are expected to vanish from the Dutch pig population in the near future, accuracy of prediction was evaluated for different scenarios representing a wide range of different proportions for entire males, castrated males and females in the Dutch pig population. The prediction equations for the instruments are in compliance with the EC regulations for prediction accuracy for the different scenarios. So, these equations will remain valid when castrated males are (gradually) removed from the Dutch slaughter population. Results of this study are of interest for researchers from countries or areas contemplating the use of one of the aforementioned instruments. The statistical approach for evaluation of prediction accuracy is of particular interest when changes in proportions of important subpopulations in the target population are foreseen.