Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Noninvasive Prediction of Beef Longissimus Tenderness

Authors
item Wheeler, Tommy
item Shackelford, Steven
item King, David
item Koohmaraie, Mohammad

Submitted to: American Meat Science Association Conference Reciprocal Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: April 15, 2007
Publication Date: June 20, 2007
Citation: Wheeler, T.L., Shackelford, S.D., King, D.A., Koohmaraie, M. 2007. Noninvasive prediction of beef longissimus tenderness. Proc. 60th Reciprocal Meat Conference, June 17-20, 2007, Brookings, SD. http://www.meatscience.org/pubs/rmcarchv/2007/presentations/RMC_2007_060_4_084_Wheeler.pdf

Technical Abstract: Previously, we developed accurate technology to predict tenderness of U.S. Select beef longissimus using visible and near-infrared reflectance spectroscopy (VISNIR). Originally this technology was applied on the bloom chain in commercial packing plants at approximately 2 min after the carcasses were ribbed. It would greatly facilitate implementation of this technology if it could be used at any point in the conventional grading process. Preliminary data indicated that variation in bloom time would cause considerable bias in existing prediction models. Therefore, a strategy was employed to develop models for prediction of tenderness that would not be subject to such bias. Spectra were collected for U.S. Select carcasses (n = 1,155) in three packing plants. Each carcass was evaluated with one instrument soon after ribbing and with another instrument when it arrived at the grading stand. At 14 d postmortem, longissimus slice shear force was measured. Overall, 4.9% of samples had SSF values > 25 kg. Data from one-half of the carcasses (n = 579) from each plant was used to develop a partial least squares regression model to predict SSF from VISNIR spectra. Data from the other carcasses (n = 576) was used to validate the model. Of the carcasses in the validation data set that were classified as tender by the VISNIR system, 0.9% had SSF values > 25 kg. Subsequently, this model was independently tested in four packing plants using U.S. Select and U.S. Choice carcasses (n = 1,973). In the independent test, spectra were collected and analyzed using a single instrument located on or near the grading stand. Overall, 16.3% of samples had SSF values > 25 kg at 14 d postmortem. Of the carcasses that were classified as tender (n = 900) by the VISNIR system, 6.8% had SSF values > 25 kg. These experiments show that tenderness of longissimus steaks can be predicted non-invasively using VISNIR spectroscopy. Application of this technology will facilitate identification of carcasses with superior tenderness for branded beef programs.

Last Modified: 9/20/2014
Footer Content Back to Top of Page