Authentication of barley-finished beef using visible and near infrared spectroscopy (Vis-NIRS) and different discrimination approaches.

Affiliation

Barragán W(1), Aalhus JL(2), Penner G(3), Dugan MER(2), Juárez M(2), López-Campos Ó(2), Vahmani P(4), Segura J(2), Angulo J(5), Prieto N(6).
Author information:
(1)Agriculture and Agri-Food Canada, Lacombe Research and Development Centre, Lacombe, Alberta T4L 1W1, Canada; Faculty of Agricultural Sciences, University of Antioquia, Carrera 75 #65-87, Medellín, Colombia; Corporación Colombiana de Investigación Agropecuaria
(AGROSAVIA), Turipaná Research Centre, Km 13 road Montería - Cereté, Córdoba, Colombia.
(2)Agriculture and Agri-Food Canada, Lacombe Research and Development Centre, Lacombe, Alberta T4L 1W1, Canada.
(3)College of Agriculture and Bioresources, University of Saskatchewan, Saskatoon, SK S7N 5A8, Canada.
(4)Department of Animal Science, University of California, 2201 Meyer Hall, Davis, CA 95616, United States.
(5)Faculty of Agricultural Sciences, University of Antioquia, Carrera 75 #65-87, Medellín, Colombia.
(6)Agriculture and Agri-Food Canada, Lacombe Research and Development Centre, Lacombe, Alberta T4L 1W1, Canada. Electronic address: [Email]

Abstract

This study evaluated visible and near-infrared spectroscopy (Vis-NIRS) to authenticate barley-finished beef using different discrimination approaches. Dietary grain source (barley, corn, or blend-50% barley/50% corn) did not affect (P > 0.05) meat quality but influenced (P < 0.05) fatty acid profiles. The longissimus thoracis (LT) from barley-fed steers had lower n-6 fatty acid content and n-6/n-3 ratio compared to LT from corn and blended grain-fed steers. Vis-NIRS coupled with partial least square discriminant analysis (PLS-DA) and support vector machine in the linear (L-SVM) kernel classified with approximately 70% overall accuracy subcutaneous fat and intact LT samples, respectively, from barley, corn, and blended-fed cattle. When only barley and corn samples were considered, fat and intact LT samples were correctly classified with overall accuracy >94% with PLS-DA and radial/L-SVM, and approximately 90% with PLS-DA and L-SVM, respectively. Ground LT samples were classified with ≤70% overall accuracy. Vis-NIRS measurements on fat and intact LT have potential to discriminate between corn and barley-fed beef.