In the Meat Standards Australia (MSA) and Guaranteed Global Grading (3G) grading schemes, beef marbling is scored visually in the chiller by accredited graders from 100 to 1190 marble score points in increments of 10. This study aimed to evaluate a hand-held camera (Q-FOM™ Beef) for determining MSA marbling scores of carcasses quartered between the 5th and 6th rib. The carcasses were scored by two accredited graders, including an expert grader (i.e. a more experienced grader). The R of correlation between scores of the two graders for 377 carcasses was 0.78 with a RMSE of 47.9. The R of correlation between the scores of the expert grader and the Q-FOM™ for 285 carcasses was 0.75 with a RMSE of 44.9. For the grader-to-grader comparison, 75.9 %, 97.1 % and 100 % of the values were within 50, 100 and 200 marbling points, respectively. For the comparison between Q-FOM™ predictions and the expert grader, 78.6 %, 96.8 % and 99.7 % of the values were within 50, 100 and 200 marbling points, respectively. Both between visual graders and the Q-FOM™ against expert grader showed acceptable accuracy performance and fulfilled the accreditation criteria defined by AUS-MEAT in Australia. Additionally, 124 Q-FOM™ images were assessed on-screen by the expert grader. The R of correlation between the in-chiller and on-screen MSA marbling scores was 0.78 with a RMSEP of 48.7. Thus, on-screen assessing met requirements for accreditation, and both in-chiller and on-screen visual assessments of MSA marbling score are acceptable inputs for developing a Q-FOM™ Beef marbling model.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1016/j.meatsci.2025.109759 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!