Systems for determining carcass lean meat yield% in beef and lamb
Gardner, G., Anderson, F., Williams, A., Ball, A.J., Hancock, B. and Pethick, D.W. (2012) Systems for determining carcass lean meat yield% in beef and lamb. In: 63rd Annual Meeting of the European Federation of Animal Science, 27 - 31 August, Bratislava, Slovakia.
Carcass lean meat weld percentage (LMY%) is a key profit driver for beef and lamb processors. Currently LMY% is poorly assessed in Australia, as processors rely on carcass weight and a single point measure of fatness to estimate IÆ4Y%. This approach has a limited capacity to predict carcass LMY%, and almost no capacity to describe variation in the distribution of lean between regions of the carcass. In response to this the Australian lamb industry has been assessing and developing a number of different high-throughput technologies to predict LMY% within abattoirs. This activity is being driven at two levels, the first targeting simpler and less expensive devices that will deliver single-site measurements of tissue depth. These devices predict LMY% with less precision (R2 range from 0.2-0.4), and at the whole carcass level only. They include mechanical tissue depth probes, ultrasound, and boning room vision systems that can determine tissue depth at the GR site (11 cm from mid-line over the 12th rib), or muscle and fat depth at the C-site (5 cm from the mid-line over the 12th rib). Secondly we are assessing more expensive whole carcass systems that will enable more accurate determination of LMY% (R2 range from 0.4-0.7) as well as determining lean distribution between different regions of the carcass. These include carcass vision systems, dual energy x-ray absosptiometry, and computer aided tomography scanning (CTscan). In all cases these LMY% prediction devices are trained upon a central ‘gold-standard’ dataset generated using CTscan of carcasses scanned in 3 sections (fore, saddle and hind). This central CTscan dataset has the advantage of generating consistent and repeatable data, not subject to human bias. Thus processors can select an LMY% prediction technology that best optimises their trade-off between cost speed, and precision. This model is now being adapted to the beef industry with obvious constraints being the size and expense of working with beef carcasses.
|Publication Type:||Conference Item|
|Murdoch Affiliation:||School of Veterinary and Biomedical Sciences|
|Item Control Page|