Report Detail Page
Lamb Eating Quality and Supply Chain Grading
1. A prototype DEXA system has been installed at the JBS abattoir in Bordertown, SA, mirroring the system installed at the Finegand abattoir, near Balclutha, NZ. Work was undertaken to establish the algorithms to predict carcase composition in lambs.
2. Experiment 1 demonstrated that the prototype DEXA can effectively determine chemical fat % and tissue depth and remove bone-containing tissue from images based upon DEXA R-values.
3. Using the relationships established in Experiment 1, Experiment 2 then tested the DEXA system across an initial group of 48 lamb carcases randomly chosen from one days' production, but selected across a diverse range of weight (17-32kg hot carcase weight) and fatness (5-27mm GR tissue depth). This demonstrated excellent precision for determining carcase fat % using computed tomography (CT) (R2 = 0.84, RMSE = 1.60). However, the association with CT lean% and bone % were less precise (CT lean % R2 = 0.64, RMSE = 1.89; CT bone % R2 = 0.64, RMSE = 0.90).
4. Experiment 3 then tested the DEXA system across a much larger population of lambs. These were from MLA's nucleus flock, with 600 lambs re-located to South Australia, feedlot finished to target slaughter weights, and then killed in groups of about 100 lambs, with each group balanced for sire. This produced a population of lambs that were spread across a diverse range of fatness (2-44mm GR tissue depth), weight (10.9-39.3kg hot carcase weight), and genotype. The final DEXA prediction equation was established within this population, and tested for robustness by transporting the prediction equation between slaughter groups. The precision of these models was very high, with R2 for CT fat %, lean % and bone % of 0.89, 0.74, 0.71, and root mean square error of 1.42, 1.69, and 0.80, which represented 6%, 9%, and 8% of the data range across nucleus flocks 2-6 for CT fat, lean and bone. This precision was maintained when these equations were derived within one slaughter group and transported to the others, and the accuracy for predicting CT fat % was also maintained within 1%.
5. Experiment 4 tested the precision of prediction for CT composition within regions. For fat% and lean% it was reduced when predicted within the fore, saddle and hind sections using DEXA values specific to these regions. Alternatively, when using the whole carcase DEXA value to predict CT composition the precision was diminished by markedly less in the fore and hind quarter, and within the saddle section it was the same or even slightly improved (for bone and lean) compared to the whole carcase predictions of composition.
6. These results demonstrate that this system will provide precise and accurate prediction of carcase composition enabling more accurate valuation of carcasses up and down the supply chain on the basis of lean meat yield.
This page was last updated on 21/06/2017
Join myMLA today
One username and password for key integrity and information Systems (LPA/NVD, NLIS, MSA & LDL).
A personalised online dashboard that provides news, weather, events and R&D tools relevant to you.
Customised market information and analysis.
Already registered for myMLA?