Hoffman Louwrens Christiaan, Wu Wencong, Zhang Shuxin, Beya Michel, Cozzolino Daniel
Centre for Nutrition and Food Sciences (CNAFS), Queensland Alliance for Agriculture and Food Innovation (QAAFI), The University of Queensland, Brisbane, QLD 4072, Australia.
Foods. 2025 Apr 21;14(8):1430. doi: 10.3390/foods14081430.
This study aimed to evaluate the utilisation of near-infrared (NIR) spectroscopy combined with chemometric techniques to identify the addition of goat liver to goat minced meat and to monitor the shelf life of the samples up to 8 days of storage. Mix samples were created by adding goat liver to goat meat in different ratios (0%, 2%, 4%, 6%, and 8% w/w), and after mincing, the samples were stored under chilled (2-4 °C) conditions for 8 days. The NIR spectra, CIELab parameters, and pH of the mixture samples were collected at the start of the study and after 2, 4, 6, and 8 days of storage. The mince became darker with the increase in days of storage, while the pH value was not affected by days of storage. Partial least squares (PLS) regression was used to develop calibration models for the CIELab parameters to predict the level of liver addition to minced meat and to predict days of storage. The standard error in cross-validation (SECV) and the coefficient of determination in cross-validation (R) were 0.10 (SECV: 3.3), 0.63 (SECV: 1.5), and 0.60 (SECV: 0.90) for L*, a*, and b*, respectively. The R and SECV were 0.32 (SECV: 2.4%) and 0.92 (SECV: 0.98 days) to predict the level of liver addition to minced meat and days of storage, respectively. The NIR calibration models developed to predict the CIELab parameters and level of addition of liver to minced meat were inadequate for predicting new samples. On the other hand, the PLS models developed could predict the days of storage, R 0.92 (SECV: 0.98 days). Compared with traditional methods such as CIELab or pH measurements, NIR spectroscopy can yield results more rapidly. However, the variability in the data set should be increased to allow the development of more reliable models.
本研究旨在评估近红外（NIR）光谱结合化学计量技术在识别山羊肉末中添加山羊肝脏以及监测样品长达8天储存期的货架期方面的应用。通过以不同比例（0%、2%、4%、6%和8% w/w）向山羊肉中添加山羊肝脏来制备混合样品，切碎后，将样品在冷藏（2 - 4°C）条件下储存8天。在研究开始时以及储存2、4、6和8天后收集混合样品的近红外光谱、CIELab参数和pH值。随着储存天数的增加，肉末颜色变深，而pH值不受储存天数的影响。采用偏最小二乘法（PLS）回归为CIELab参数建立校准模型，以预测肉末中肝脏的添加水平并预测储存天数。L*、a和b的交叉验证标准误差（SECV）和交叉验证决定系数（R）分别为0.10（SECV：3.3）、0.63（SECV：1.5）和0.60（SECV：0.90）。预测肉末中肝脏添加水平和储存天数的R和SECV分别为0.32（SECV：2.4%）和0.92（SECV：0.98天）。为预测CIELab参数和肉末中肝脏添加水平而建立的近红外校准模型在预测新样品时不够充分。另一方面，所建立的PLS模型可以预测储存天数，R为0.92（SECV：0.98天）。与CIELab或pH测量等传统方法相比，近红外光谱可以更快地得出结果。然而，应增加数据集的变异性，以开发更可靠的模型。
