The beef industry has a considerable need for an improved measure of saleable red meat yield (RMY) for cattle and beef carcasses. Technologies with the ability to more directly measure or estimate RMY need to be identified and developed so that appropriate market signals can be sent to producers to improve the efficiency, profitability and sustainability of the beef industry. Recent research has demonstrated a discrepancy between USDA Yield Grade (YG) and subprimal yield (SY). The current YG equation utilizes only four predictors acquired from a cross-sectional area of the carcass between the 12th and 13th rib. Including additional independent variables may enhance the ability to forecast SY and carcass value. Since the 1990’s, 3D imaging technologies have been advancing rapidly. This includes the development of software platforms that can integrate images from multiple cameras to simultaneously and instantaneously build a true 360-degree, 3D image. The nature of 3D imaging is such that every image is a product of dimensional measurements that exist as data in the background. These data, albeit very numerous, can be utilized with modern machine learning, artificial intelligence statistical models to estimate or predict object volumes, and, in the case of beef carcasses and primals, composition by incorporating other factors such as carcass or primal weight. X-ray technologies, including real-time whole body composition scanners, specifically iDXA that has demonstrated the capacity to determine whole body composition (fat, muscle, and bone) and bone density in human medicine applications, research trials studying the development of domesticated animals, including cattle and pigs, and in meat and livestock industry applications in other countries, are viable options for online application in today’s commercial industry. The purpose of this study is to demonstrate a proof of concept for 3D imaging and iDXA for determining the RMY if beef carcasses and/or primals, by performing an in-house carcass and primal yield test coupled with compositional measurements of lean, fat, and bone.
This study utilized a combination of 3D imaging and iDXA x-ray data collection to evaluate the composition of carcasses and primal sections in beef cattle. At the Texas Tech meat laboratory, a grand total of 41 cattle, comprising of steers, heifers, and bulls, were processed according to the established norms of the industry. Following the harvest, the carcasses underwent 3D imaging using LIDAR technology, while the primal quarters were additionally scanned using iDXA technology. Subsequently, the carcasses were processed into boneless subprimal cuts, and the weights of different components were meticulously documented using certified scales. Data management encompasses the computation of carcass recovery weight (CSW) and the classification of components into subprimal, fat, bone, and trim categories. The lean trim underwent a conversion into modified red meat and modified fat. The cross-sectional measurements of the 3D images were obtained by processing them using MeshLab, while the composition results and linear measurements were obtained from iDXA scans. The user employed multiple linear regression (MLR) to perform statistical analyses and develop predictive models for red meat yield, adjusted fat percentage, and bone percentage.
The results demonstrate extremely precise predictive models, with 96% of the variation in carcass red meat yield, 97% in adjusted fat percentage, and 95% in bone percentage being explained. Comparable achievements were noted in individual primal cuts, as the models effectively accounted for the significant variations in red meat yield, adjusted fat percentage, and bone percentage for chuck, rib, loin, and round cuts. Moreover, iDXA scans have shown efficacy in forecasting fat composition, specifically in the loin and rib primals. Nonetheless, the capacity to predict red meat yield and bone composition differed across various primal cuts, with greater precision observed in the loin and rib cuts as compared to the chuck and round cuts.
Ultimately, the combination of 3D imaging and iDXA technology, along with rigorous statistical analyses, offers a thorough method for precisely forecasting the composition of carcasses and primal cuts in beef cattle. These findings have ramifications for enhancing efficacy and accuracy in meat processing and supply chain management within the beef industry.