Prediction of live and carcass characteristics of market hogs by use of a single longitudinal ultrasonic scan.

Live animal and carcass data were collected from market barrows and gilts (n = 119) to determine the accuracy and precision of using a single longitudinal scan, parallel to midline, in estimating body and carcass composition. Data from test pigs (n = 96) were used to develop prediction equations. Best results were obtained in predicting weight of boneless cuts (ham, loin, and shoulder), weight of lean cuts (bone-in ham, loin, and shoulder), and weight of fat-standardized carcass lean. Less accuracy was obtained in predicting ratios of all estimates on a live basis. Independent variables analyzed for the live models were live weight, sex, and ultrasonic fat depth and muscle depth at the 10th rib. Independent variables for the carcass models were the same as on the live animal; the estimators for boneless cuts (ham, loin, and shoulder) were most precise. Equations were tested against an independent set of experimental pigs (n = 23). Equations for predicting weight of boneless cuts, weight of ham and loin, and percentage of fat-standardized lean using both live and carcass measurements were most accurate, with R2 values between .78 and .87 and RSD values between 1.30 and 1.92 kg. The results of this study reinforce the potential of assessing carcass composition and value by using a single longitudinal B-mode scan on both live pigs and carcass; live weight, sex of pig, and fat depth at the 10th rib were the greatest contributors to variation.