Lean content prediction in pig carcasses, loin and ham by computed tomography (CT) using a density model

Area of Science:

• Animal Science
• Agricultural Engineering
• Food Science

Background:

• Accurate prediction of carcass weight and lean content is crucial for the pork industry.
• Current methods may be labor-intensive or lack precision.
• Standardized and efficient assessment tools are needed to comply with EU regulations.

Purpose of the Study:

• To develop and validate a computed tomography (CT) based predictive model for pig carcass weight and lean content.
• To assess the accuracy of the CT model against manual dissection.
• To provide a simple, accurate method for pork quality assessment.

Main Methods:

• Computed tomography (CT) scanning of 122 Spanish pig half carcasses, 52 hams, and 52 loins.
• Development of predictive models based on histogram analysis and a density correction equation.
• Correlation of CT model predictions with manual dissection data using root mean square error of calibration (RMSEC).

Main Results:

• The CT model achieved a root mean square error of calibration (RMSEC) of 0.6 kg for half carcass weight.
• Lean meat percentage prediction yielded RMSEC values of 1.48% for carcasses, 0.97% for hams, and 1.07% for loins.
• The developed methodology demonstrated high predictive accuracy for both weight and lean content.

Conclusions:

• Computed tomography offers a simple and accurate method for predicting pig carcass weight and lean content.
• The CT-based model meets the accuracy requirements for compliance with EU regulations.
• This technology can enhance efficiency and standardization in the pork industry.