Abstract The aim of the present work was to evaluate the ability of a portable near-infrared (NIR) spectrometer to detect adulteration in ground meat. NIR spectroscopy has been used in food science for 40 years. Miniaturization, however, is a recent approach that allows in situ analysis. In this study, samples of meat were adulterated in the range of 0–100 wt% in binary blends (chicken/beef; beef/pork; pork/chicken) and ternary blends (beef/chicken/pork). For the binary blends, values of R2c and R2p ranged from 0.78 to 0.99. Optimal results were achieved to predict the chicken content in beef blends (R2c = 0.98; R2p = 0.99; RMSEC = 4.5 wt%; RMSEP =3.5 wt%; LOD = 3.4 wt% and LOQ 11.2 wt%). For the ternary blends, the analytical performance was considered to be good only for the prediction of beef content, with the following values: R2c = 0.98, R2p = 0.93, RMSEC = 3.6 wt%, RMSEP =4.7 wt%, LOD =4.7 wt% and LOQ 15.7 wt%. Finally, in order to improve the analytical performance of the regression models developed thus far, the NIR spectra of the binary and ternary blends were grouped into a unique matrix. PLS and SVR regression models were constructed, and the SVR model resulted in performance parameters superior to the PLS. Nevertheless, the portable NIR spectrometer showed satisfactory performance for the quantification of beef in ground meat blends (chicken/beef, pork/beef, and chicken/beef/pork).

中文翻译：

---

使用便携式 NIR 光谱仪定量碎肉中的牛肉、猪肉和鸡肉

摘要 本工作的目的是评估便携式近红外 (NIR) 光谱仪检测碎肉掺假的能力。近红外光谱已在食品科学中使用了 40 年。然而，小型化是一种允许原位分析的最新方法。在本研究中，肉类样品在二元混合物（鸡肉/牛肉；牛肉/猪肉；猪肉/鸡肉）和三元混合物（牛肉/鸡肉/猪肉）中掺假了 0–100 wt%。对于二元混合物，R2c 和 R2p 的值范围从 0.78 到 0.99。实现了预测牛肉混合物中鸡肉含量的最佳结果（R2c = 0.98；R2p = 0.99；RMSEC = 4.5 wt%；RMSEP = 3.5 wt%；LOD = 3.4 wt% 和 LOQ 11.2 wt%）。对于三元混合物，分析性能被认为仅适用于预测牛肉含量，具有以下值：R2c = 0.98，R2p = 0.93，RMSEC = 3.6 wt%，RMSEP = 4.7 wt%，LOD = 4.7 wt% 和 LOQ 15.7 wt%。最后，为了提高迄今为止开发的回归模型的分析性能，将二元和三元混合物的 NIR 光谱分组到一个独特的矩阵中。构建了PLS和SVR回归模型，SVR模型得到的性能参数优于PLS。尽管如此，便携式 NIR 光谱仪对碎肉混合物（鸡肉/牛肉、猪肉/牛肉和鸡肉/牛肉/猪肉）中牛肉的定量表现出令人满意的性能。构建了PLS和SVR回归模型，SVR模型得到的性能参数优于PLS。尽管如此，便携式 NIR 光谱仪对碎肉混合物（鸡肉/牛肉、猪肉/牛肉和鸡肉/牛肉/猪肉）中牛肉的定量表现令人满意。构建了PLS和SVR回归模型，SVR模型得到的性能参数优于PLS。尽管如此，便携式 NIR 光谱仪对碎肉混合物（鸡肉/牛肉、猪肉/牛肉和鸡肉/牛肉/猪肉）中牛肉的定量表现出令人满意的性能。