Prediction of cooking loss of pork belly using quality properties of pork loin

The predictability of cooking loss in pork belly using the quality properties of pork loin was investigated. Pork belly at the 6th thoracic vertebra and pork loin at the 14th thoracic vertebra from 120 pork carcasses were used in this study. Quality properties, such as pH, proximate composition, color (L*, a*, and b* values), and cooking loss were measured. Linear regression analysis showed that the L*, a*, and b* values of pork loin were significant variables for predicting the cooking loss of pork belly (P < 0.05). However, the adjusted correlation coefficient (R2) of the linear regression was 0.51. Logistic regression analysis for the prediction of cooking loss groups (low, middle, and high) of pork belly, with the L*, a*, and b* values as the independent variables, resulted in 84% concordance. Pork carcasses can be sorted based on the cooking loss groups of pork belly by using the color parameters of pork loin.