Alamprese Cristina, Amigo José Manuel, Casiraghi Ernestina, Engelsen Søren Balling
Department of Food, Environmental and Nutritional Sciences (DeFENS), Università degli Studi di Milano, via Celoria 2, 20133 Milano, Italy.
Department of Food Science, Spectroscopy and Chemometrics, Faculty of Science, University of Copenhagen, Rolighedsvej 26, DK-1958 Frederiksberg C, Denmark.
Meat Sci. 2016 Nov;121:175-181. doi: 10.1016/j.meatsci.2016.06.018. Epub 2016 Jun 16.
This work aims at the development of a method based on FT-NIR spectroscopy and multivariate analysis for the identification and quantification of minced beef meat adulteration with turkey meat. Samples were analyzed as raw, frozen-thawed and cooked. Different multivariate regression and class-modeling strategies were evaluated. PLS regression models with R(2) in prediction higher than 0.884 and RMSEP lower than 10.8% were developed. PLS-DA applied to discriminate each type of sample in two classes (adulteration threshold=20%) showed values of sensitivity and specificity in prediction higher than 0.84 and 0.76, respectively. Thus, the study demonstrates that FT-NIR spectroscopy coupled with suitable chemometric strategies is a reliable tool for the identification and quantification of minced beef adulteration with turkey meat not only in fresh products, but also in frozen-thawed and cooked samples. This achievement is of crucial importance in the meat industry due to the increasing number of processed meat products, in which technological treatments can mask a possible inter-species adulteration.
本研究旨在开发一种基于傅里叶变换近红外光谱（FT-NIR）和多变量分析的方法，用于鉴定和定量牛肉末中掺假的火鸡肉。对样品进行了生肉、冻融和熟肉分析。评估了不同的多变量回归和分类建模策略。建立了预测相关系数（R(2)）高于0.884且预测均方根误差（RMSEP）低于10.8%的偏最小二乘（PLS）回归模型。应用偏最小二乘判别分析（PLS-DA）将每种类型的样品分为两类（掺假阈值 = 20%），预测的灵敏度和特异性值分别高于0.84和0.76。因此，该研究表明，FT-NIR光谱结合合适的化学计量学策略是一种可靠的工具，不仅可用于鉴定和定量新鲜产品中牛肉末掺假的火鸡肉，还可用于冻融和熟肉样品。由于加工肉类产品数量不断增加，而其中的技术处理可能掩盖种间掺假情况，这一成果在肉类行业至关重要。
