Chemometrics and hyperspectral imaging applied to assessment of chemical, textural and structural characteristics of meat

Validation of a handheld near-infrared spectrophotometer for measurement of chemical intramuscular fat in Australian lamb

The objective of the study was to independently validate a calibrated commercial handheld near infrared (NIR) spectroscopic device and test its repeatability over time using phenotypically diverse populations of Australian lamb. Validation testing in eight separate data sub-groups (n = 1591 carcasses overall) demonstrated that the NIR device had moderate precision (R2 = 0.4-0.64, RMSEP = 0.70-1.22%) but fluctuated in accuracy between experimental site demonstrated by variable slopes (0.50-0.94) and biases (-0.86-0.02). The repeatability experiment (n = 10 carcasses) showed that time to scan post quartering affected NIR measurement from 0 to 24 h (P < 0.001). On average, NIR IMF% was 0.97% lower (P < 0.001) at 24 h (4.01% ± 0.166), compared to 0 h. There was no difference (P > 0.05) between Time 0 and 1 h or Time 0 and 4 h or between replicate scans within each time point. This study demonstrated the SOMA NIR device could predict lamb chemical IMF% with moderate precision and accuracy, however additional work is required to understand how loin preparation, blooming and surface hydration affect NIR measurement.

Hyperspectral imaging through vacuum packaging for monitoring cheese biochemical transformation caused by Clostridium metabolism

This study developed a novel method for monitoring cheese contamination with Clostridium spores non-invasively using hyperspectral imaging (HSI). The ability of HSI to quantify Clostridium metabolites was investigated with control cheese and cheese manufactured with milk contaminated with Clostridium tyrobutyricum, Clostridium butyricum and Clostridium sporogenes. Microbial count, HSI and SPME-GC-MS data were obtained over 10 weeks of storage. The developed method using HSI successfully quantified butyric acid (R² = 0.91, RPD = 3.38) a major compound of Clostridium metabolism in cheese. This study creates a new venue to monitor the spatial and temporal development of late blowing defect (LBD) in cheese using fast and non-invasive measurement.

Nondestructive detection of nutritional parameters of pork based on NIR hyperspectral imaging technique

Nondestructive detection of the nutritional parameters of pork is of great importance. This study aimed to investigate the feasibility of applying hyperspectral image technology to detect the nutrient content and distribution of pork nondestructively. Hyperspectral cubes of 100 pork samples were collected using a line-scan hyperspectral system, the effects of different preprocessing methods on the modeling effects were compared and analyzed, the feature wavelengths of fat and protein were extracted, and the full-wavelength model was optimized using the regressor chains (RC) algorithm. Finally, pork's fat, protein, and energy value distributions were visualized using the best prediction model. The results showed that standard normal variate was more effective than other preprocessing methods, the feature wavelengths extracted by the competitive adaptive reweighted sampling algorithm had better prediction performance, and the protein model prediction performance was optimized after using the RC algorithm. The best prediction models were developed, with the correlation coefficient of prediction (R_P) = 0.929, the root mean square error in prediction (RMSEP) = 0.699% and residual prediction deviation (RPD) = 2.669 for fat, and R_P = 0.934, RMSEP = 0.603% and RPD = 2.586 for protein. The pseudo-color maps were helpful for the analysis of nutrient distribution in pork. Hyperspectral image technology can be a fast, nondestructive, and accurate tool for quantifying the composition and assessing the distribution of nutrients in pork.

Adulteration detection in minced beef using low-cost color imaging system coupled with deep neural network

Major processed meat products, including minced beef, are one of the favorite ingredients of most people because they are high in protein, vitamins, and minerals. The high demand and high prices make processed meat products vulnerable to adulteration. In addition, eliminating morphological attributes makes the authenticity of minced beef challenging to identify with the naked eye. This paper aims to describe the feasibility study of adulteration detection in minced beef using a low-cost imaging system coupled with a deep neural network. The proposed method was expected to be able to detect minced beef adulteration. There were 500 captured images of minced beef samples. Then, there were 24 color and textural features retrieved from the image. The samples were then labeled and evaluated. A deep neural network (DNN) was developed and investigated to support classification. The proposed DNN was also compared to six machine learning algorithms in the form of accuracy, precision, and sensitivity of classification. The feature importance analysis was also performed to obtain the most impacted features to classification results. The DNN model classification accuracy was 98.00% without feature selection and 99.33% with feature selection. The proposed DNN has the best performance with individual accuracy of up to 99.33%, a precision of up to 98.68%, and a sensitivity of up to 98.67%. This work shows the enormous potential application of a low-cost imaging system coupled with DNN to rapidly detect adulterants in minced beef with high performance.

Hyperspectral imaging and deep learning for quantification of Clostridium sporogenes spores in food products using 1D- convolutional neural networks and random forest model

Clostridium sporogenes spores are used as surrogates for Clostridium botulinum, to verify thermal exposure and lethality in sterilization regimes by food industries. Conventional methods to detect spores are time-consuming and labour intensive. The objectives of this study were to evaluate the feasibility of using hyperspectral imaging (HSI) and deep learning approaches, firstly to identify dead and live forms of C. sporogenes spores and secondly, to estimate the concentration of spores on culture media plates and ready-to-eat mashed potato (food matrix). C. sporogenes spores were inoculated by either spread plating or drop plating on sheep blood agar (SBA) and tryptic soy agar (TSA) plates and by spread plating on the surface of mashed potato. Reflectance in the spectral range of 547-1701 nm from the region of interest was used for principal component analysis (PCA). PCA was successful in distinguishing dead and live spores and different levels of inoculum (10² to 10⁶ CFU/ml) on both TSA and SBA plates, however, was not efficient on the mashed potato (food matrix). Hence, deep learning classification frameworks namely 1D- convolutional neural networks (CNN) and random forest (RF) model were used. CNN model outperformed the RF model and the accuracy for quantification of spores was improved by 4% and 8% in the presence and absence, respectively of dead spores. The screening system used in this study was a combination of HSI and deep learning modelling, which resulted in an overall accuracy of 90-94% when the dead/inactivated spores were present and absent, respectively. The only discrepancy detected was during the prediction of samples with low inoculum levels (<10² CFU/ml). In summary, it was evident that HSI in combination with a deep learning approach showed immense potential as a tool to detect and quantify spores on nutrient media as well as on specific food matrix (mashed potato). However, the presence of dead spores in any sample is postulated to affect the accuracy and would need replicates and confirmatory assays.

Aspects of the influence of vegetable-oil-based marinade on organoleptic and physicochemical indicators of the quality of semi-finished natural marinated meat products

This work evidences the expediency of marinade application based on blends of refined vegetable oils in the technology of natural marinated semi-finished meat products. Formulations of natural marinated semi-finished meat products using blends of refined vegetable oils (rapeseed, olive, and sunflower) enriched with the enzyme bromelain were developed in this work. Such formulations can provide the human body with the necessary amount of protein, released by enzymatic hydrolysis of the connective tissue proteins collagen and elastin catalyzed by the plant-based enzyme bromelain, as well as the necessary ω-3 and ω-6 polyunsaturated fatty acids, which are not synthesized in the human body but sourced from food and are one of the main structural units for many vital processes. The article presents the results of a study of organoleptic and physicochemical parameters of natural marinated semi-finished meat products from beef and pork. It is confirmed that the use of rapeseed oil and rapeseed:sunflower oil blend in a ratio of 70:30 produce the best organoleptic characteristics in both semi-finished beef and semi-finished pork: a tender, juicy texture, a pleasant taste and aroma, and an attractive appearance. It has been proven that the use of marinades based on blends of refined vegetable oils reduces the free moisture content and moisture-retaining capacity of the product, due to the presence of refined vegetable oils in the marinade, which contributes to the binding of moisture. The use of marinades based on refined vegetable oils can increase product yield. Vegetable oil, when used in meat marinade, mitigates the ill effects of acid inclusion. It dissolves well the aromas of added spices, and as it seeps through the structure of the meat, the oil gently envelops it, sealing in moisture and preventing drying out during cooking.

Three-dimensional tissue fabrication system by co-culture of microalgae and animal cells for production of thicker and healthy cultured food

OBJECTIVES

"Cultured food" is focused worldwide as "the third stage in meat production system" after hunting and livestock farming, and a sustainable food production system. In this study, we attempted to fabricate a three-dimensional (3-D) tissue by co-cultivation of animal cells with photosynthetic autotrophic microalgae so as to produce thicker and healthy cultured foods.

RESULTS

Metabolism and damage of co-cultured tissues fabricated by microalgae, Chlorella vulgaris (C. vulgaris), and C2C12 cells were compared to monoculture tissues fabricated by C2C12 animal cells alone. Although the metabolism of monoculture tissue showed anaerobic respiration (ratio of lactate production to glucose consumption, LG ratio: 2.01 ± 0.15), that of the co-culture tissue partially changed to efficient aerobic respiration (LG ratio: 1.58 ± 0.14). In addition, the amount of ammonia in the culture media decreased markedly by co-cultivation. The release of lactate dehydrogenase from the thicker tissue was one-seventh in the co-cultivation, showing improved tissue damage. The co-cultivation with microalgae improved the culture condition of thicker tissues, resulting in the fabrication/maintenance of 200-400 µm-thickness tissues. The co-cultured tissue fabricated by microalgae and animal cells was not only rich in nutrients but also enabled thicker tissue fabrication without tissue damage as compared to tissue fabricated by animal cells alone.

CONCLUSIONS

This tissue fabrication system by co-culture of microalgae and animal cells will be a valuable tool for the production of thicker and healthy cultured food.

Identification of Weeds Based on Hyperspectral Imaging and Machine Learning

Weeds can be major environmental and economic burdens in New Zealand. Traditional methods of weed control including manual and chemical approaches can be time consuming and costly. Some chemical herbicides may have negative environmental and human health impacts. One of the proposed important steps for providing alternatives to these traditional approaches is the automated identification and mapping of weeds. We used hyperspectral imaging data and machine learning to explore the possibility of fast, accurate and automated discrimination of weeds in pastures where ryegrass and clovers are the sown species. Hyperspectral images from two grasses (Setaria pumila [yellow bristle grass] and Stipa arundinacea [wind grass]) and two broad leaf weed species (Ranunculus acris [giant buttercup] and Cirsium arvense [Californian thistle]) were acquired and pre-processed using the standard normal variate method. We trained three classification models, namely partial least squares-discriminant analysis, support vector machine, and Multilayer Perceptron (MLP) using whole plant averaged (Av) spectra and superpixels (Sp) averaged spectra from each weed sample. All three classification models showed repeatable identification of four weeds using both Av and Sp spectra with a range of overall accuracy of 70-100%. However, MLP based on the Sp method produced the most reliable and robust prediction result (89.1% accuracy). Four significant spectral regions were found as highly informative for characterizing the four weed species and could form the basis for a rapid and efficient methodology for identifying weeds in ryegrass/clover pastures.

Rheological and Antimicrobial Properties of Chitosan and Quinoa Protein Filmogenic Suspensions with Thyme and Rosemary Essential Oils

Food packaging faces the negative impact of synthetic materials on the environment, and edible coatings offer one alternative from filmogenic suspensions (FS). In this work, an active edible FS based on chitosan (C) and quinoa protein (QP) cross-linked with transglutaminase was produced. Thyme (T) and rosemary (R) essential oils (EOs) were incorporated as antimicrobial agents. Particle size, Z potential, and rheological parameters were evaluated. The antimicrobial activity against Micrococcus luteus (NCIB 8166) and Salmonella sp. (Lignieres 1900) was monitored using atomic force microscopy and image analysis. Results indicate that EOs incorporation into C:QP suspensions did not affect the Z potential, ranging from -46.69 ± 3.19 mV to -46.21 ± 3.83 mV. However, the polydispersity index increased from 0.51 ± 0.07 to 0.80 ± 0.04 in suspensions with EO. The minimum inhibitory concentration of active suspensions against Salmonella sp. was 0.5% (v/v) for thyme and 1% (v/v) for rosemary. Entropy and fractal dimension of the images were used to confirm the antimicrobial effect of EOs, which modified the surface roughness.

Myoglobin-Based Classification of Minced Meat Using Hyperspectral Imaging

Minced meat substitution is one of the most common frauds which not only affects consumer health but impacts their lifestyles and religious customs as well. A number of methods have been proposed to overcome these frauds; however, these mostly rely on laboratory measures and are often subject to human error. Therefore, this study proposes novel hyperspectral imaging (400–1000 nm) based non-destructive isos-bestic myoglobin (Mb) spectral features for minced meat classification. A total of 60 minced meat spectral cubes were pre-processed using true-color image formulation to extract regions of interest, which were further normalized using the Savitzky–Golay filtering technique. The proposed pipeline outperformed several state-of-the-art methods by achieving an average accuracy of 88.88%.

Hyperspectral Imaging Coupled with Multivariate Analysis and Image Processing for Detection and Visualisation of Colour in Cooked Sausages Stuffed in Different Modified Casings

A hyperspectral imaging system was for the first time exploited to estimate the core colour of sausages stuffed in natural hog casings or in two hog casings treated with solutions containing surfactants and lactic acid in slush salt. Yellowness of sausages stuffed in natural hog casings (control group, 20.26 ± 4.81) was significantly higher than that of sausages stuffed in casings modified by submersion for 90 min in a solution containing 1:30 (w/w) soy lecithin:distilled water, 2.5% wt. soy oil, and 21 mL lactic acid per kg NaCl (17.66 ± 2.89) (p < 0.05). When predicting the lightness and redness of the sausage core, a partial least squares regression model developed from spectra pre-treated with a second derivative showed calibration coefficients of determination (R_c²) of 0.73 and 0.76, respectively. Ten, ten, and seven wavelengths were selected as the important optimal wavelengths for lightness, redness, and yellowness, respectively. Those wavelengths provide meaningful information for developing a simple, cost-effective multispectral system to rapidly differentiate sausages based on their core colour. According to the canonical discriminant analysis, lightness possessed the highest discriminant power with which to differentiate sausages stuffed in different casings.

NIR Hyperspectral Imaging Technology Combined with Multivariate Methods to Identify Shrimp Freshness

In this study, a hyperspectral imaging system of 866.4–1701.0 nm, combined with a variety of spectral processing methods were adopted to identify shrimp freshness. To gain the optimal model combination, three preprocessing methods (Savitzky-Golay first derivative (SG1), multivariate scatter correction (MSC), and standard normal variate (SNV)), three characteristic wavelength extraction algorithms (random frog algorithm (RFA), uninformative variables elimination (UVE), and competitive adaptive reweighted sampling (CARS)), and four discriminant models (partial least squares discrimination analysis (PLS-DA), least squares support vector machine (LSSVM), random forest (RF), and extreme learning machine (ELM)) were employed for experimental study. First of all, due to the full wavelength modeling analysis, three preprocessing methods were utilized to preprocess the original spectral data. The analysis showed that the spectral data processed by the SNV method had the best performance among the four discriminant models. Secondly, due to the characteristic wavelength modeling analysis, three characteristic wavelength extraction algorithms were utilized to extract the characteristic wavelength of the SNV-processed spectral data. It was found that the CARS algorithm achieved the best performance among the three characteristic wavelength extraction algorithms, and the combining adoption of the ELM model and different characteristic wavelength extraction algorithms obtained the best results. Therefore, the model based on SNV-CARS-ELM obtained the best performance and was elected as the optimal model. Lastly, for accurately and explicitly displaying the refrigeration days of shrimps, the original hyperspectral images of shrimps were substituted into the SNV-CARS-ELM model, thus obtaining the general classification accuracy of 97.92%, and the object-wise method was used to visualize the classification results. As a result, the method proposed in this study can effectively detect the freshness of shrimps.

Identification of Cold Spots Using Non-Destructive Hyperspectral Imaging Technology in Model Food Processed by Coaxially Induced Microwave Pasteurization and Sterilization

The model food in this study known as mashed potato consisted of ribose (1.0%) and lysine (0.5%) to induce browning via Maillard reaction products. Mashed potato was processed by Coaxially Induced Microwave Pasteurization and Sterilization (CiMPAS) regime to generate an F0 of 6-8 min and analysis of the post-processed food was done in two ways, which included by measuring the color changes and using hyperspectral data acquisition. For visualizing the spectra of each tray in comparison with the control sample (raw mashed-potato), the mean spectrum (i.e., mean of region of interest) of each tray, as well as the control sample, was extracted and then fed to the fitted principal component analysis model and the results coincided with those post hoc analysis of the average reflectance values. Despite the presence of a visual difference in browning, the Lightness (L) values were not significantly (p < 0.05) different to detect a cold spot among a range of 12 processed samples. At the same time, hyperspectral imaging could identify the colder trays among the 12 samples from one batch of microwave sterilization.

Comparative review and the recent progress in detection technologies of meat product adulteration

Meat adulteration, mainly for the purpose of economic pursuit, is widespread and leads to serious public health risks, religious violations, and moral loss. Rapid, effective, accurate, and reliable detection technologies are keys to effectively supervising meat adulteration. Considering the importance and rapid advances in meat adulteration detection technologies, a comprehensive review to summarize the recent progress in this area and to suggest directions for future progress is beneficial. In this review, destructive meat adulteration technologies based on DNA, protein, and metabolite analyses and nondestructive technologies based on spectroscopy were comparatively analyzed. The advantages and disadvantages, application situations of these technologies were discussed. In the future, determining suitable indicators or markers is particularly important for destructive methods. To improve sensitivity and save time, new interdisciplinary technologies, such as biochips and biosensors, are promising for application in the future. For nondestructive techniques, convenient and effective chemometric models are crucial, and the development of portable devices based on these technologies for onsite monitoring is a future trend. Moreover, omics technologies, especially proteomics, are important methods in laboratory detection because they enable multispecies detection and unknown target screening by using mass spectrometry databases.

Combination of spectra and texture data of hyperspectral imaging for prediction and visualization of palmitic acid and oleic acid contents in lamb meat

The feasibility of combining spectral and textural information from hyperspectral imaging to improve the prediction of the C16:0 and C18:1 n9 contents for lamb was explored. 29 and 22 optimal wavelengths were selected for the C16:0 and C18:1 n9 contents, respectively, by conducting the variable combination population analysis-iteratively retaining informative variables (VCPA-IRIV) algorithm. To extract the textural features of images, a gray-level co-occurrence matrix (GLCM) analysis was implemented in the first principal component image. The least squares support vector machine (LSSVM) model and the partial least squares regression (PLSR) model were developed to predict the C16:0 and C18:1 n9 contents from the spectra and the fusion data. The distribution map was visualized using the best model with the imaging process. The results showed that the combination of the spectral and textural information of hyperspectral imaging coupled with the VCPA-IRIV algorithm had strong potential for the prediction and visualization of the C16:0 and C18:1 n9 contents of lamb.

Application of Hyperspectral Imaging as a Nondestructive Technique for Foodborne Pathogen Detection and Characterization

Microbial food safety is a persistent and exacting global issue due to the multiplicity and complexity of foods and food production systems. Foodborne illnesses caused by foodborne bacterial pathogens frequently occur, thus endangering the safety and health of human beings. Factors such as pretreatments, that is, culturing, enrichment, amplification make the traditional routine identification and enumeration of large numbers of bacteria in a complex microbial consortium complex, expensive, and time-consuming. Therefore, the need for rapid point-of-use detection systems for foodborne bacterial pathogens with high sensitivity and specificity is crucial in food safety control. Hyperspectral imaging (HSI) as a powerful testing technology provides a rapid, nondestructive approach for pathogen detection. This article reviews some fundamental information about HSI, including instrumentation, data acquisition, image processing, and data analysis-the current application of HSI for the detection, classification, and discrimination of various foodborne pathogens. The merits and demerits of HSI for pathogen detection as well as current and future trends are discussed. Therefore, the purpose of this review is to provide a brief overview of HSI, and further lay emphasis on the emerging trend and importance of this technique for foodborne pathogen detection.

Deep Learning Meets Hyperspectral Image Analysis: A Multidisciplinary Review

Modern hyperspectral imaging systems produce huge datasets potentially conveying a great abundance of information; such a resource, however, poses many challenges in the analysis and interpretation of these data. Deep learning approaches certainly offer a great variety of opportunities for solving classical imaging tasks and also for approaching new stimulating problems in the spatial-spectral domain. This is fundamental in the driving sector of Remote Sensing where hyperspectral technology was born and has mostly developed, but it is perhaps even more true in the multitude of current and evolving application sectors that involve these imaging technologies. The present review develops on two fronts: on the one hand, it is aimed at domain professionals who want to have an updated overview on how hyperspectral acquisition techniques can combine with deep learning architectures to solve specific tasks in different application fields. On the other hand, we want to target the machine learning and computer vision experts by giving them a picture of how deep learning technologies are applied to hyperspectral data from a multidisciplinary perspective. The presence of these two viewpoints and the inclusion of application fields other than Remote Sensing are the original contributions of this review, which also highlights some potentialities and critical issues related to the observed development trends.