Search Results (8)

To achieve accurate detection the content of multiple parts pork adulterated in mutton under the effect of mutton flavor essence and colorant by RGB images, the improved CBAM-Invert-ResNet50 network based on the attention mechanism and the inversion residual was used to detect the content of pork from the back, front leg, and hind leg in adulterated mutton. The deep features of different parts extracted by the CBAM-Invert-ResNet50 were fused by feature, stitched, and combined with transfer learning, and the content of pork from mixed parts in adulterated mutton was detected. The results showed that the R² of the CBAM-Invert-ResNet50 for the back, front leg, and hind leg datasets were 0.9373, 0.8876, and 0.9055, respectively, and the RMSE values were 0.0268 g·g⁻¹, 0.0378 g·g⁻¹, and 0.0316 g·g⁻¹, respectively. The R² and RMSE of the mixed dataset were 0.9264 and 0.0290 g·g⁻¹, respectively. When the features of different parts were fused, the R² and RMSE of the CBAM-Invert-ResNet50 for the mixed dataset were 0.9589 and 0.0220 g·g⁻¹, respectively. Compared with the model built before feature fusion, the R² of the mixed dataset increased by 0.0325, and the RMSE decreased by 0.0070 g·g⁻¹. The above results indicated that the CBAM-Invert-ResNet50 model could effectively detect the content of pork from different parts in adulterated mutton as additives. Feature fusion combined with transfer learning can effectively improve the detection accuracy for the content of mixed parts of pork in adulterated mutton. The results of this study can provide technical support and a basis for maintaining the mutton market order and protecting mutton food safety supervision. Full article

Food adulteration is a serious problem all over the world. Establishing an accurate, sensitive and fast detection method is an important part of identifying food adulteration. Herein, a sequence-specific ladder-shape melting temperature isothermal amplification (LMTIA) assay was reported to detect soybean-derived components using proofreading enzyme-mediated probe cleavage (named Proofman), which could realize real-time and visual detection without uncapping. The results showed that, under the optimal temperature of 57 °C, the established Proofman–LMTIA method for the detection of soybean-derived components in dairy products was sensitive to 1 pg/μL, with strong specificity, and could distinguish soybean genes from those of beef, mutton, sunflower, corn, walnut, etc. The established Proofman–LMTIA detection method was applied to the detection of actual samples of cow milk and goat milk. The results showed that the method was accurate, stable and reliable, and the detection results were not affected by a complex matrix without false positives or false negatives. It was proved that the method could be used for the detection and identification of soybean-derived components in actual dairy products samples. Full article

To realize the real-time automatic identification of adulterated minced mutton, a convolutional neural network (CNN) image recognition model of adulterated minced mutton was constructed. Images of mutton, duck, pork and chicken meat pieces, as well as prepared mutton adulterated with different proportions of duck, pork and chicken meat samples, were acquired by the laboratory’s self-built image acquisition system. Among all images were 960 images of different animal species and 1200 images of minced mutton adulterated with duck, pork and chicken. Additionally, 300 images of pure mutton and mutton adulterated with duck, pork and chicken were reacquired again for external validation. This study compared and analyzed the modeling effectiveness of six CNN models, AlexNet, GoogLeNet, ResNet-18, DarkNet-19, SqueezeNet and VGG-16, for different livestock and poultry meat pieces and adulterated mutton shape feature recognition. The results show that ResNet-18, GoogLeNet and DarkNet-19 models have the best learning effect and can identify different livestock and poultry meat pieces and adulterated minced mutton images more accurately, and the training accuracy of all three models reached more than 94%, among which the external validation accuracy of the optimal three models for adulterated minced mutton images reached more than 70%. Image learning based on a deep convolutional neural network (DCNN) model can identify different livestock meat pieces and adulterated mutton, providing technical support for the rapid and nondestructive identification of mutton authenticity. Full article

In order to seek high profit, businesses mix beef and mutton with cheap meat, such as duck, pork, and chicken. Five pairs of primers were designed for quintuple droplet digital PCR (qddPCR) of specific genomic regions from five selected species and specificity and amplification efficiency were determined. The mixed DNA template with an equal copy number was used for detecting the accuracy and limit of multiplex PCR. The results showed that the primers and probes of the five selected species had good specificity with the minimum number of detection copies: 0.15 copies/µL beef (Bos taurus), 0.28 copies/μL duck (Anas platyrhynchos), 0.37 copies/μL pork (Sus scrofa), 0.39 copies/μL chicken (Gallus gallus), and 0.41 copies/μL mutton (Ovis aries), respectively. The five sets of primers and probes could quickly judge whether the specified meat components existed in the food commodities. Full article

Single-probe near-infrared spectroscopy (NIRS) usually uses different spectral information for modelling, but there are few reports about its influence on model performance. Based on sized-adaptive online NIRS information and the 2D conventional neural network (CNN), minced samples of pure mutton, pork, duck, and adulterated mutton with pork/duck were classified in this study. The influence of spectral information, convolution kernel sizes, and classifiers on model performance was separately explored. The results showed that spectral information had a great influence on model accuracy, of which the maximum difference could reach up to 12.06% for the same validation set. The convolution kernel sizes and classifiers had little effect on model accuracy but had significant influence on classification speed. For all datasets, the accuracy of the CNN model with mean spectral information per direction, extreme learning machine (ELM) classifier, and 7 × 7 convolution kernel was higher than 99.56%. Considering the rapidity and practicality, this study provides a fast and accurate method for online classification of adulterated mutton. Full article

The evaluation of mutton adulteration faces new challenges because of mutton flavour essence, which achieves a similar flavour between the adulterant and mutton. Hence, methods for classifying and quantifying the adulterated mutton under the effect of mutton flavour essence, based on near-infrared hyperspectral imaging (NIR-HSI, 1000–2500 nm) combined with machine learning (ML) and sparrow search algorithm (SSA), were proposed in this study. After spectral preprocessing via first derivative combined with multiple scattering correction (1D + MSC), classification and quantification models were established using back propagation neural network (BP), extreme learning machine (ELM) and support vector machine/regression (SVM/SVR). SSA was further used to explore the global optimal parameters of these models. Results showed that the performance of models improves after optimisation via the SSA. SSA-SVM achieved the optimal discrimination result, with an accuracy of 99.79% in the prediction set; SSA-SVR achieved the optimal prediction result, with an R_P² of 0.9304 and an RMSEP of 0.0458 g·g⁻¹. Hence, NIR-HSI combined with ML and SSA is feasible for classification and quantification of mutton adulteration under the effect of mutton flavour essence. This study can provide a theoretical and practical reference for the evaluation and supervision of food quality under complex conditions. Full article

To make meat adulteration detection systems faster, simpler and more efficient, we established a duck-derived meat rapid detection Recombinase Polymerase Amplification (dRPA) method by using interleukin 2 (IL-2) from nuclear genomic DNA as the target gene to design specific primers. We tested the dRPA detection system by comparing its sensitivity and specificity using real-time fluorescent PCR technology. By adjusting the ratio of reagents, this method shortens the time of DNA extraction and visualizes results in combination with colloidal gold immunoassay strips. Our results demonstrate that this dRPA method could specifically detect duck-derived components with a sensitivity of up to 23 copies/μL and the accuracy of the results is consistent with real-time fluorescent PCR. Additionally, dRPA can detect at least 1% of the duck meat content by mixing beef and mutton with duck meat in different proportions, which was verified by spot-check market samples. These results can be visualized with colloidal gold immunoassay strips with the same accuracy as real-time fluorescent RPA. dRPA can complete detection within 30 min, which shortens existing detection time and quickly visualizes the detection results on-site. This lays the groundwork for future large-scale standardized duck origin detection. Full article

Processed meat is a target in meat adulteration for economic gain. This study demonstrates a molecular and bioinformatics diagnostic pipeline, utilizing the mitochondrial 16S ribosomal RNA (rRNA) gene, to determine processed meat product mislabeling through Next-Generation Sequencing. Nine pure meat samples were collected and artificially mixed at different ratios to verify the specificity and sensitivity of the pipeline. Processed meat products (n = 155), namely, minced meat, biltong, burger patties, and sausages, were collected across South Africa. Sequencing was performed using the Illumina MiSeq sequencing platform. Each sample had paired-end reads with a length of ±300 bp. Quality control and filtering was performed using BBDuk (version 37.90a). Each sample had an average of 134,000 reads aligned to the mitochondrial genomes using BBMap v37.90. All species in the artificial DNA mixtures were detected. Processed meat samples had reads that mapped to the Bos (90% and above) genus, with traces of reads mapping to Sus and Ovis (2–5%) genus. Sausage samples showed the highest level of contamination with 46% of the samples having mixtures of beef, pork, or mutton in one sample. This method can be used to authenticate meat products, investigate, and manage any form of mislabeling. Full article