Prediction of Loin Area by Depth of Loin in Pigs

Abstract

This study examined the relationship between real loin area and loin depth of a carcass that had been measured using image processing of carcass cross-sectional photographs. Thereby, we established a method of estimating the loin cross-sectional area from loin depth measured in live pigs using an ultrasonic device. First, we examined the relation between loin area (REM1/2) and loin depth at the position of a half body length. At that position, loin depth was measured at 2 cm, 4 cm, 6 cm, and 8 cm side from the midline (P2, P4, P6, and P8, respectively). Next, we examined the relations between the loin area (REMLAST) and the loin depth at the position of the last rib (LP4, LP6, LP8). The first examination used 149 Duroc pigs (103 castrates and 46 gilts). The second examination assessed 220 Duroc pigs (162 castrates and 58 gilts). Six trapezoid areas were calculated from the loin depth at the position of half body length : ((1) S1=(P2+P4)×2/2, (2) S2=(P4+P6)×2/2, (3) S3=(P6+P8)×2/2, (4) S4=S1+S2=P2+2P4+P6, (5) S5=S2+S3=P4+2P6+P8, (6) S6=S1+S2+S3=P2+2P4+2P6+P8). The coefficient of correlation between the carcass real loin area (REM1/2) and the live cross-sectional area (SEM1/2), as measured using an ultrasonic color machine, was 0.730 (p<0.001). The relationship between real loin area (REM1/2) and the real loin depth at 6 cm (P6) was highest (0.817, p<0.001) for the carcass loin at half body length. The relation with S5 (0.855, P<0.001) was the highest in six areas. Multiple regression analysis, of which the independent variables are depth and six areas, indicated this equation with the highest adjusted contribution (R²=0.764) : REM1/2=2.125S5-0.730P8-4.883. For the carcass loin area at the last rib, the relationship between the loin area and the depth at 8 cm from the midline (LP8) was highest (0.830, p<0.001) ; those with LP4 and LP6 were 0.604 (p<0.001) and 0.795 (p<0.001), respectively. When three depths of the last rib (LP4, LP6, LP8) were assumed as independent variables, the regression equation of EMLAST=1.104LP4+1.551LP6+3.757LP8+1.118, and an adjusted contribution of 0.724 was obtained. A regression equation of EMLAST=6.012LP6+0.913 and an adjusted contribution of 0.631 were obtained when we assumed LP6 as the independent variable considering live pig measurements.