Submitted to: International Congress of Meat Science and Technology Proceedings
Publication Type: Proceedings
Publication Acceptance Date: August 28, 1995
Publication Date: N/A
The objective of this study was to compare the efficacy of electromagnetic scanning to USDA yield grade for predicting subprimal weight and yield, and carcass value for beef. Crossbred steers (n=219) of diverse genotypes were fed and serially slaughtered on four dates at 21 d intervals. The calculated yield grade from yield grade factors and the USDA grader yield grade were obtained at 24 h postmortem. A hindquarter of each animal was scanned in a pork carcass electromagnetic scanner at 2.5 MHz. Each side was processed into primal and subprimal cuts (.76 and 0 cm of fat trim), bone, and fat. Prices for calculating value were from November, 1994. Electromagnetic scanning predicted lean weight in the round, loin, rib, and chuck at .76 cm fat trim with coefficient of determination (CD) and root mean square error (RMSE) OF .91 (1.0 kg), .82 (.8 kg), .78 (.6 kg), .81 (1.7 kg), respectively. Accuracy of prediction for primals at 0 cm fat trim and subprimals at either trim level were similar to that for primals at .76 cm fat trim. Total side lean predicted from scan peak, hindquarter weight, and hindquarter length had CD and RMSE of .92 (3.5 kg) and .91 (3.6 kg) for .76 and 0 cm fat trim, respectively. Equations including electromagnetic scanning were slightly better than either method of determining yield grade for predicting lean weight (CD = .92 vs .91 vs .90). When predicting value per kg carcass weight, calculated yield grade was more accurate than equations including electromagnetic scanning. These data indicate that electromagnetic scanning can accurately predict lean weight, however, it does not improve over USDA yield grade for predicting yield of lean or value per kg carcass weight.