A machine learning workflow for raw food spectroscopic classification in a future industry

Predicting multiple taste sensations with a multiobjective machine learning method

Taste perception plays a pivotal role in guiding nutrient intake and aiding in the avoidance of potentially harmful substances through five basic tastes - sweet, bitter, umami, salty, and sour. Taste perception originates from molecular interactions in the oral cavity between taste receptors and chemical tastants. Hence, the recognition of taste receptors and the subsequent perception of taste heavily rely on the physicochemical properties of food ingredients. In recent years, several advances have been made towards the development of machine learning-based algorithms to classify chemical compounds' tastes using their molecular structures. Despite the great efforts, there remains significant room for improvement in developing multi-class models to predict the entire spectrum of basic tastes. Here, we present a multi-class predictor aimed at distinguishing bitter, sweet, and umami, from other taste sensations. The development of a multi-class taste predictor paves the way for a comprehensive understanding of the chemical attributes associated with each fundamental taste. It also opens the potential for integration into the evolving realm of multi-sensory perception, which encompasses visual, tactile, and olfactory sensations to holistically characterize flavour perception. This concept holds promise for introducing innovative methodologies in the rational design of foods, including pre-determining specific tastes and engineering complementary diets to augment traditional pharmacological treatments.

Improved organic and pesticide-free rice (Oryza sativa L.) authentication based on multiple stable isotope ratio analysis and rice milling state

This study looked at the application of multiple bulk stable isotope ratio analysis to accurately authenticate organic rice and counteract organic fraud within the expanding global organic market. Variations of δ13C, δ15N, δ18O, and δ34S in organic, pesticide-free, and conventional rice were assessed across different milling states (brown, milled, and bran). Individual stable isotope ratio alone such as δ15N demonstrated limited capacity to correctly differentiate organic, pesticide-free, and conventional rice. A support vector machine model-incorporating δ13C, δ15N, δ18O, and δ34S in milled rice-yielded overall predictability (95%) in distinguishing organic, pesticide-free, and conventional rice, where δ18O emerged as the pivotal variable based on the feature weights in the SVM model. These findings suggest the potential of multi-isotope and advanced statistical approaches in combating organic fraud and ensuring authenticity in the food supply chain.

Real-Time AI-Assisted Push-Broom Hyperspectral System for Precision Agriculture

In the ever-evolving landscape of modern agriculture, the integration of advanced technologies has become indispensable for optimizing crop management and ensuring sustainable food production. This paper presents the development and implementation of a real-time AI-assisted push-broom hyperspectral system for plant identification. The push-broom hyperspectral technique, coupled with artificial intelligence, offers unprecedented detail and accuracy in crop monitoring. This paper details the design and construction of the spectrometer, including optical assembly and system integration. The real-time acquisition and classification system, utilizing an embedded computing solution, is also described. The calibration and resolution analysis demonstrates the accuracy of the system in capturing spectral data. As a test, the system was applied to the classification of plant leaves. The AI algorithm based on neural networks allows for the continuous analysis of hyperspectral data relative up to 720 ground positions at 50 fps.

Geographical and varietal origin differentiation of alcoholic beverages through the association between FT-Raman spectroscopy and advanced data processing strategies

The present work aimed to test the efficiency of FT-Raman spectroscopy for fruit spirits discrimination by developing differentiation models based on two approaches, namely a supervised statistical method (Partial Least Squares Discriminant Analysis), and a Machine Learning technique (Support Vector Machines). For this purpose, a data set comprising 86 Romanian distillate samples was used, which aimed to be differentiated in terms of the raw material used for production (plum, apple, pear and grape) and county of origin (Cluj, Satu Mare and Salaj). Eight distinct preprocessing methods (autoscale, mean center, variance scaling, smoothing, 1st derivative, 2nd derivative, standard normal variate and Pareto) followed by a feature selection step were applied to identify the meaningful input data based on which the most efficient classification models can be constructed. Both types of models led to accuracy scores greater than 90% in differentiating the distillate samples in terms of geographical and botanical origin.

Deep learning framework for sensor array precision and accuracy enhancement

In the upcoming years, artificial intelligence is going to transform the practice of medicine in most of its specialties. Deep learning can help achieve better and earlier problem detection, while reducing errors on diagnosis. By feeding a deep neural network (DNN) with the data from a low-cost and low-accuracy sensor array, we demonstrate that it becomes possible to significantly improve the measurements' precision and accuracy. The data collection is done with an array composed of 32 temperature sensors, including 16 analog and 16 digital sensors. All sensors have accuracies between [Formula: see text]. 800 vectors are extracted, covering a range from to 30 to [Formula: see text]. In order to improve the temperature readings, we use machine learning to perform a linear regression analysis through a DNN. In an attempt to minimize the model's complexity in order to eventually run inferences locally, the network with the best results involves only three layers using the hyperbolic tangent activation function and the Adam Stochastic Gradient Descent optimizer. The model is trained with a randomly-selected dataset using 640 vectors (80% of the data) and tested with 160 vectors (20%). Using the mean squared error as a loss function between the data and the model's prediction, we achieve a loss of only 1.47x10[Formula: see text] on the training set and 1.22x10[Formula: see text] on the test set. As such, we believe this appealing approach offers a new pathway towards significantly better datasets using readily-available ultra low-cost sensors.

Preliminary investigation into the prediction of indicators of beef spoilage using Raman and Fourier transform infrared spectroscopy

The objective of this study was to assess the potential to predict the microbial beef spoilage indicators by Raman and Fourier transform infrared (FT-IR) spectroscopies. Vacuum skin packaged (VSP) beef steaks were stored at 0 °C, 4 °C, 8 °C and under a dynamic temperature condition (0 °C ∼ 4 °C ∼ 8 °C, for 36 d). Total viable count (TVC) and total volatile basic nitrogen (TVB-N) were obtained during the storage period along with spectroscopic data. The Raman and FTIR spectra were baseline corrected, pre-processed using Savitzky-Golay smoothing and normalized. Subsequently partial least squares regression (PLSR) models of TVC and TVB-N were developed and evaluated. The root mean squared error (RMSE) ranged from 0.81 to1.59 (log CFU/g or mg/100 g) and the determination coefficient (R²) from 0.54 to 0.75. The performance of PLSR model based on data fusion (combination of Raman and FT-IR data) is better than that based on Raman spectra and similar to that of FT-IR. Overall, Raman spectroscopy, FT-IR spectroscopy, and a combination of both exhibited a potential for the prediction of the beef spoilage.

A robust and resilience machine learning for forecasting agri-food production

This research proposes a new framework for agri-food capacity production by considering resiliency and robustness and paying attention to disruption and risk for the first time. It is applied robust stochastic optimization by adding robustness to the constraint's objective function and resiliency situation. This research minimizes the mean absolute deviation and coefficient of standard deviation errors by linear function in the agri-food capacity production. This study suggests agri-food managers and decision-makers use this mathematical method to forecast and improve production management. The results of this research lead to better decision-making and are compared with other sine functions. The main model's Robust and Resiliency Mean Absolute Deviation (RRMAD) value is 1.28% lower than other sine-type functions. The conservativity coefficient, confidence level, weight factor, resiliency coefficient, and probability of the scenario vary. The main model's RRMAD value is 1.28% lower than other sine-type functions. Growing the weight factor will result in an increase in RRMAD and a smooth decline in R-squared. Additionally, as the resilience coefficient rises, the RRMAD function increases while the R-squared declines. By altering the probability of the scenario, the RRMAD function drops, and the R-squared goes up.

Assessment of the Microbial Spoilage and Quality of Marinated Chicken Souvlaki through Spectroscopic and Biomimetic Sensors and Data Fusion

Fourier-transform infrared spectroscopy (FT-IR), multispectral imaging (MSI), and an electronic nose (E-nose) were implemented individually and in combination in an attempt to investigate and, hence, identify the complexity of the phenomenon of spoilage in poultry. For this purpose, marinated chicken souvlaki samples were subjected to storage experiments (isothermal conditions: 0, 5, and 10 °C; dynamic temperature conditions: 12 h at 0 °C, 8 h at 5 °C, and 4 h at 10 °C) under aerobic conditions. At pre-determined intervals, samples were microbiologically analyzed for the enumeration of total viable counts (TVCs) and Pseudomonas spp., while, in parallel, FT-IR, MSI, and E-nose measurements were acquired. Quantitative models of partial least squares–Regression (PLS-R) and support vector machine–regression (SVM-R) (separately for each sensor and in combination) were developed and validated for the estimation of TVCs in marinated chicken souvlaki. Furthermore, classification models of linear discriminant analysis (LDA), linear support vector machine (LSVM), and cubic support vector machines (CSVM) that classified samples into two quality classes (non-spoiled or spoiled) were optimized and evaluated. The model performance was assessed with data obtained by six different analysts and three different batches of marinated souvlaki. Concerning the estimation of the TVCs via the PLS-R model, the most efficient prediction was obtained with spectral data from MSI (root mean squared error—RMSE: 0.998 log CFU/g), as well as with combined data from FT-IR/MSI (RMSE: 0.983 log CFU/g). From the developed SVM-R models, the predictions derived from MSI and FT-IR/MSI data accurately estimated the TVCs with RMSE values of 0.973 and 0.999 log CFU/g, respectively. For the two-class models, the combined data from the FT-IR/MSI instruments analyzed with the CSVM algorithm provided an overall accuracy of 87.5%, followed by the MSI spectral data analyzed with LSVM, with an overall accuracy of 80%. The abovementioned findings highlighted the efficacy of these non-invasive rapid methods when used individually and in combination for the assessment of spoilage in marinated chicken products regardless of the impact of the analyst, season, or batch.

Rapid Assessment of Microbial Quality in Edible Seaweeds Using Sensor Techniques Based on Spectroscopy, Imaging Analysis and Sensors Mimicking Human Senses

The expansion of the seaweed aquaculture sector along with the rapid deterioration of these products escalates the importance of implementing rapid, real-time techniques for their quality assessment. Seaweed samples originating from Scotland and Ireland were stored under various temperature conditions for specific time intervals. Microbiological analysis was performed throughout storage to assess the total viable counts (TVC), while in parallel FT-IR spectroscopy, multispectral imaging (MSI) and electronic nose (e-nose) analyses were conducted. Machine learning models (partial least square regression (PLS-R)) were developed to assess any correlations between sensor and microbiological data. Microbial counts ranged from 1.8 to 9.5 log CFU/g, while the microbial growth rate was affected by origin, harvest year and storage temperature. The models developed using FT-IR data indicated a good prediction performance on the external test dataset. The model developed by combining data from both origins resulted in satisfactory prediction performance, exhibiting enhanced robustness from being origin unaware towards microbiological population prediction. The results of the model developed with the MSI data indicated a relatively good prediction performance on the external test dataset in spite of the high RMSE values, whereas while using e-nose data from both MI and SAMS, a poor prediction performance of the model was reported.

Geographical origin discrimination of Agaricus bisporus produced by the complete medium: A pilot study in South Korea

The complete medium used for mushroom cultivation is important for reliable crop production. We aimed to identify how the origin of Agaricus bisporus grown in Korea was affected by complete media produced in different geographical regions using stable isotope ratios (SIRs). We found that the SIR features of A. bisporus significantly depended on the complete media origin used; in particular, it appeared the high δ³⁴S in the Chinese complete medium, low δ³⁴S in the Dutch complete medium, and high δ¹⁵N in the Korean complete medium (P < 0.05). The support vector machine method appeared better geo-origin classification of A. bisporus by the complete media compared to a linear discriminant analysis. A large-scale study should be conducted to establish a reliable origin identification model for A. bisporus grown in complete media to improve the global mushroom marketplace.

Spectroscopic Data for the Rapid Assessment of Microbiological Quality of Chicken Burgers

The rapid assessment of the microbiological quality of highly perishable food commodities is of great importance. Spectroscopic data coupled with machine learning methods have been investigated intensively in recent years, because of their rapid, non-destructive, eco-friendly qualities and their potential to be used on-, in- or at-line. In the present study, the microbiological quality of chicken burgers was evaluated using Fourier transform infrared (FTIR) spectroscopy and multispectral imaging (MSI) in tandem with machine learning algorithms. Six independent batches were purchased from a food industry and stored at 0, 4, and 8 °C. At regular time intervals (specifically every 24 h), duplicate samples were subjected to microbiological analysis, FTIR measurements, and MSI sampling. The samples (n = 274) acquired during the data collection were classified into three microbiological quality groups: “satisfactory”: 4−7 log CFU/g, “acceptable”: 7−8 log CFU/g, and “unacceptable”: >8 logCFU/g. Subsequently, classification models were trained and tested (external validation) with several machine learning approaches, namely partial least squares discriminant analysis (PLSDA), support vector machine (SVM), random forest (RF), logistic regression (LR), and ordinal logistic regression (OLR). Accuracy scores were attained for the external validation, exhibiting FTIR data values in the range of 79.41−89.71%, and, for the MSI data, in the range of 74.63−85.07%. The performance of the models showed merit in terms of the microbiological quality assessment of chicken burgers.

Truths and myths about superfoods in the era of the COVID-19 pandemic

Nowadays, during the current COVID-19 pandemic, consumers increasingly seek foods that not only fulfill the basic need (i.e., satisfying hunger) but also enhance human health and well-being. As a result, more attention has been given to some kinds of foods, termed "superfoods," making big claims about their richness in valuable nutrients and bioactive compounds as well as their capability to prevent illness, reinforcing the human immune system, and improve overall health.This review is an attempt to uncover truths and myths about superfoods by giving examples of the most popular foods (e.g., berries, pomegranates, watermelon, olive, green tea, several seeds and nuts, honey, salmon, and camel milk, among many others) that are commonly reported as having unique nutritional, nutraceutical, and functional characteristics.While superfoods have become a popular buzzword in blog articles and social media posts, scientific publications are still relatively marginal. The reviewed findings show that COVID-19 has become a significant driver for superfoods consumption. Food Industry 4.0 innovations have revolutionized many sectors of food technologies, including the manufacturing of functional foods, offering new opportunities to improve the sensory and nutritional quality of such foods. Although many food products have been considered superfoods and intensively sought by consumers, scientific evidence for their beneficial effectiveness and their "superpower" are yet to be provided. Therefore, more research and collaboration between researchers, industry, consumers, and policymakers are still needed to differentiate facts from marketing gimmicks and promote human health and nutrition.

Spectroscopy Approaches for Food Safety Applications: Improving Data Efficiency Using Active Learning and Semi-supervised Learning

The past decade witnessed rapid development in the measurement and monitoring technologies for food science. Among these technologies, spectroscopy has been widely used for the analysis of food quality, safety, and nutritional properties. Due to the complexity of food systems and the lack of comprehensive predictive models, rapid and simple measurements to predict complex properties in food systems are largely missing. Machine Learning (ML) has shown great potential to improve the classification and prediction of these properties. However, the barriers to collecting large datasets for ML applications still persists. In this paper, we explore different approaches of data annotation and model training to improve data efficiency for ML applications. Specifically, we leverage Active Learning (AL) and Semi-Supervised Learning (SSL) and investigate four approaches: baseline passive learning, AL, SSL, and a hybrid of AL and SSL. To evaluate these approaches, we collect two spectroscopy datasets: predicting plasma dosage and detecting foodborne pathogen. Our experimental results show that, compared to the de facto passive learning approach, advanced approaches (AL, SSL, and the hybrid) can greatly reduce the number of labeled samples, with some cases decreasing the number of labeled samples by more than half.

Spectroscopy and imaging technologies coupled with machine learning for the assessment of the microbiological spoilage associated to ready-to-eat leafy vegetables

Based on both new and previously utilized experimental data, the present study provides a comparative assessment of sensors and machine learning approaches for evaluating the microbiological spoilage of ready-to-eat leafy vegetables (baby spinach and rocket). Fourier-transform infrared (FTIR), near-infrared (NIR), visible (VIS) spectroscopy and multispectral imaging (MSI) were used. Two data partitioning approaches and two algorithms, namely partial least squares regression and support vector regression (SVR), were evaluated. Concerning baby spinach, when model testing was performed on samples randomly selected, the performance was better than or similar to the one attained when testing was performed based on dynamic temperatures data, depending on the applied analytical technology. The two applied algorithms yielded similar model performances for the majority of baby spinach cases. Regarding rocket, the random data partitioning approach performed considerably better results in almost all cases of sensor/algorithm combination. Furthermore, SVR algorithm resulted in considerably or slightly better model performances for the FTIR, VIS and NIR sensors, depending on the data partitioning approach. However, PLSR algorithm provided better models for the MSI sensor. Overall, the microbiological spoilage of baby spinach was better assessed by models derived mainly from the VIS sensor, while FTIR and MSI were more suitable in rocket. According to the findings of this study, a distinct sensor and computational analysis application is needed for each vegetable type, suggesting that there is not a single combination of analytical approach/algorithm that could be applied successfully in all food products and throughout the food supply chain.

New Non-Invasive Method for the Authentication of Apple Cultivars

Food authentication is very important to protect consumers, sellers, and producers from fraud. Although several methods have been developed using a wide range of analytical techniques, most of them require sample destruction and do not allow in situ sampling or analysis, nor reliable quantification of hundreds of molecules at the same time. To overcome these limitations, we have developed and validated a new noninvasive analytical workflow for food authentication. The method uses a functionalized strip to adsorb small molecules from the surface of the food product, followed by gas chromatography–mass spectrometry analysis of the desorbed analytes. We validated the method and applied it to the classification of five different apple varieties. Molecular concentrations obtained from the analysis of 44 apples were used to identify markers for apple cultivars or, in combination with machine learning techniques, to perform cultivar classification. The overall reproducibility of the method was very good, showing a good coefficient of variation for both targeted and untargeted analysis. The approach was able to correctly classify all samples. In addition, the method was also used to detect pesticides and the following molecules were found in almost all samples: chlorpyrifos-methyl, deltamethrin, and malathion. The proposed approach not only showed very good analytical performance, but also proved to be suitable for noninvasive food authentication and pesticide residue analysis.

Application of machine learning to the monitoring and prediction of food safety: A review

Machine learning (ML) has proven to be a useful technology for data analysis and modeling in a wide variety of domains, including food science and engineering. The use of ML models for the monitoring and prediction of food safety is growing in recent years. Currently, several studies have reviewed ML applications on foodborne disease and deep learning applications on food. This article presents a literature review on ML applications for monitoring and predicting food safety. The paper summarizes and categorizes ML applications in this domain, categorizes and discusses data types used for ML modeling, and provides suggestions for data sources and input variables for future ML applications. The review is based on three scientific literature databases: Scopus, CAB Abstracts, and IEEE. It includes studies that were published in English in the period from January 1, 2011 to April 1, 2021. Results show that most studies applied Bayesian networks, Neural networks, or Support vector machines. Of the various ML models reviewed, all relevant studies showed high prediction accuracy by the validation process. Based on the ML applications, this article identifies several avenues for future studies applying ML models for the monitoring and prediction of food safety, in addition to providing suggestions for data sources and input variables.

Microbiological Quality Assessment of Chicken Thigh Fillets Using Spectroscopic Sensors and Multivariate Data Analysis

Fourier transform infrared spectroscopy (FT-IR) and multispectral imaging (MSI) were evaluated for the prediction of the microbiological quality of poultry meat via regression and classification models. Chicken thigh fillets (n = 402) were subjected to spoilage experiments at eight isothermal and two dynamic temperature profiles. Samples were analyzed microbiologically (total viable counts (TVCs) and Pseudomonas spp.), while simultaneously MSI and FT-IR spectra were acquired. The organoleptic quality of the samples was also evaluated by a sensory panel, establishing a TVC spoilage threshold at 6.99 log CFU/cm². Partial least squares regression (PLS-R) models were employed in the assessment of TVCs and Pseudomonas spp. counts on chicken’s surface. Furthermore, classification models (linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), support vector machines (SVMs), and quadratic support vector machines (QSVMs)) were developed to discriminate the samples in two quality classes (fresh vs. spoiled). PLS-R models developed on MSI data predicted TVCs and Pseudomonas spp. counts satisfactorily, with root mean squared error (RMSE) values of 0.987 and 1.215 log CFU/cm², respectively. SVM model coupled to MSI data exhibited the highest performance with an overall accuracy of 94.4%, while in the case of FT-IR, improved classification was obtained with the QDA model (overall accuracy 71.4%). These results confirm the efficacy of MSI and FT-IR as rapid methods to assess the quality in poultry products.

Data Science in the Food Industry

Food safety is one of the main challenges of the agri-food industry that is expected to be addressed in the current environment of tremendous technological progress, where consumers' lifestyles and preferences are in a constant state of flux. Food chain transparency and trust are drivers for food integrity control and for improvements in efficiency and economic growth. Similarly, the circular economy has great potential to reduce wastage and improve the efficiency of operations in multi-stakeholder ecosystems. Throughout the food chain cycle, all food commodities are exposed to multiple hazards, resulting in a high likelihood of contamination. Such biological or chemical hazards may be naturally present at any stage of food production, whether accidentally introduced or fraudulently imposed, risking consumers' health and their faith in the food industry. Nowadays, a massive amount of data is generated, not only from the next generation of food safety monitoring systems and along the entire food chain (primary production included) but also from the Internet of things, media, and other devices. These data should be used for the benefit of society, and the scientific field of data science should be a vital player in helping to make this possible.

Long-term isotopic model study for ecofriendly rice (Oryza sativa L.) authentication: Updating a case study in South Korea

To overcome the lack of consumer trust in ecofriendly products due to low reliability of ecofriendly certification and decreasing areas certified for growing ecofriendly agricultural products, alternative approaches for reliable certification are required. Isotopic-chemometric analysis has potential for determining organic authenticity, but previous studies have struggled to differentiate the authenticities of different rice types. The present study examined 5-year variations in δ¹³C and δ¹⁵N in ecofriendly and conventional rice sold at retail markets in South Korea, while assessing the feasibility of discriminant models for authentication of organic rice. Supporting vector machine analysis showed 4.4-14.6% better overall predictability of rice types than discriminant analysis and was effective in discriminating organic or conventional rice from pesticide-free rice, potentially enabling high-throughput screening to authenticate organic rice at marketplaces. Our findings provide reliable information for authenticating ecofriendly rice, with a potential to improve consumer safety and thus the confidence in organic products.

Detection of Meat Adulteration Using Spectroscopy-Based Sensors

Minced meat is a vulnerable to adulteration food commodity because species- and/or tissue-specific morphological characteristics cannot be easily identified. Hence, the economically motivated adulteration of minced meat is rather likely to be practiced. The objective of this work was to assess the potential of spectroscopy-based sensors in detecting fraudulent minced meat substitution, specifically of (i) beef with bovine offal and (ii) pork with chicken (and vice versa) both in fresh and frozen-thawed samples. For each case, meat pieces were minced and mixed so that different levels of adulteration with a 25% increment were achieved while two categories of pure meat also were considered. From each level of adulteration, six different samples were prepared. In total, 120 samples were subjected to visible (Vis) and fluorescence (Fluo) spectra and multispectral image (MSI) acquisition. Support Vector Machine classification models were developed and evaluated. The MSI-based models outperformed the ones based on the other sensors with accuracy scores varying from 87% to 100%. The Vis-based models followed in terms of accuracy with attained scores varying from 57% to 97% while the lowest performance was demonstrated by the Fluo-based models. Overall, spectroscopic data hold a considerable potential for the detection and quantification of minced meat adulteration, which, however, appears to be sensor-specific.

Assessing the Biofilm Formation Capacity of the Wine Spoilage Yeast Brettanomyces bruxellensis through FTIR Spectroscopy

Brettanomyces bruxellensis is a wine spoilage yeast known to colonize and persist in production cellars. However, knowledge on the biofilm formation capacity of B. bruxellensis remains limited. The present study investigated the biofilm formation of 11 B. bruxellensis strains on stainless steel coupons after 3 h of incubation in an aqueous solution. FTIR analysis was performed for both planktonic and attached cells, while comparison of the obtained spectra revealed chemical groups implicated in the biofilm formation process. The increased region corresponding to polysaccharides and lipids clearly discriminated the obtained spectra, while the absorption peaks at the specific wavenumbers possibly reveal the presence of β-glucans, mannas and ergosterol. Unsupervised clustering and supervised classification were employed to identify the important wavenumbers of the whole spectra. The fact that all the metabolic fingerprints of the attached versus the planktonic cells were similar within the same cell phenotype class and different between the two phenotypes, implies a clear separation of the cell phenotype; supported by the results of the developed classification model. This study represents the first to succeed at applying a non-invasive technique to reveal the metabolic fingerprint implicated in the biofilm formation capacity of B. bruxellensis, underlying the homogenous mechanism within the yeast species.

Rapid Microbial Quality Assessment of Chicken Liver Inoculated or Not With Salmonella Using FTIR Spectroscopy and Machine Learning

Chicken liver is a highly perishable meat product with a relatively short shelf-life and that can get easily contaminated with pathogenic microorganisms. This study was conducted to evaluate the behavior of spoilage microbiota and of inoculated Salmonella enterica on chicken liver. The feasibility of Fourier-transform infrared spectroscopy (FTIR) to assess chicken liver microbiological quality through the development of a machine learning workflow was also explored. Chicken liver samples [non-inoculated and inoculated with a four-strain cocktail of ca. 10³ colony-forming units (CFU)/g Salmonella] were stored aerobically under isothermal (0, 4, and 8°C) and dynamic temperature conditions. The samples were subjected to microbiological analysis with concomitant FTIR measurements. The developed FTIR spectral analysis workflow for the quantitative estimation of the different spoilage microbial groups consisted of robust data normalization, feature selection based on extra-trees algorithm and support vector machine (SVM) regression analysis. The performance of the developed models was evaluated in terms of the root mean square error (RMSE), the square of the correlation coefficient (R²), and the bias (B_f) and accuracy (A_f) factors. Spoilage was mainly driven by Pseudomonas spp., followed closely by Brochothrix thermosphacta, while lactic acid bacteria (LAB), Enterobacteriaceae, and yeast/molds remained at lower levels. Salmonella managed to survive at 0°C and dynamic conditions and increased by ca. 1.4 and 1.9 log CFU/g at 4 and 8°C, respectively, at the end of storage. The proposed models exhibited A_f and B_f between observed and predicted counts within the range of 1.071 to 1.145 and 0.995 to 1.029, respectively, while the R² and RMSE values ranged from 0.708 to 0.828 and 0.664 to 0.949 log CFU/g, respectively, depending on the microorganism and chicken liver samples. Overall, the results highlighted the ability of Salmonella not only to survive but also to grow at refrigeration temperatures and demonstrated the significant potential of FTIR technology in tandem with the proposed spectral analysis workflow for the estimation of total viable count, Pseudomonas spp., B. thermosphacta, LAB, Enterobacteriaceae, and Salmonella on chicken liver.

Spoilage assessment of chicken breast fillets by means of fourier transform infrared spectroscopy and multispectral image analysis

The objective of this research was the evaluation of Fourier transforms infrared spectroscopy (FT-IR) and multispectral image analysis (MSI) as efficient spectroscopic methods in tandem with multivariate data analysis and machine learning for the assessment of spoilage on the surface of chicken breast fillets. For this purpose, two independent storage experiments of chicken breast fillets (n ​= ​215) were conducted at 0, 5, 10, and 15 ​°C for up to 480 ​h. During storage, samples were analyzed microbiologically for the enumeration of Total Viable Counts (TVC) and Pseudomonas spp. In addition, FT-IR and MSI spectral data were collected at the same time intervals as for microbiological analyses. Multivariate data analysis was performed using two software platforms (a commercial and a publicly available developed platform) comprising several machine learning algorithms for the estimation of the TVC and Pseudomonas spp. population of the surface of the samples. The performance of the developed models was evaluated by intra batch and independent batch testing. Partial Least Squares- Regression (PLS-R) models from the commercial software predicted TVC with root mean square error (RMSE) values of 1.359 and 1.029 log CFU/cm² for MSI and FT-IR analysis, respectively. Moreover, RMSE values for Pseudomonas spp. model were 1.574 log CFU/cm² for MSI data and 1.078 log CFU/cm² for FT-IR data. From the implementation of the in-house sorfML platform, artificial neural networks (nnet) and least-angle regression (lars) were the most accurate models with the best performance in terms of RMSE values. Nnet models developed on MSI data demonstrated the lowest RMSE values (0.717 log CFU/cm²) for intra-batch testing, while lars outperformed nnet on independent batch testing with RMSE of 1.252 log CFU/cm². Furthermore, lars models excelled with the FT-IR data with RMSE of 0.904 and 0.851 log CFU/cm² in intra-batch and independent batch testing, respectively. These findings suggested that FT-IR analysis is more efficient than MSI to predict the microbiological quality on the surface of chicken breast fillets.

•

Poultry meat’s vulnerability to spoilage demands rapid quality assessment LWT-Food Sci. Technol.methods.

•

FT-IR and MSI are non-invasive methods applied in a variety of meat products.

•

SorfML is a web platform providing diverse machine learning algorithms.

•

FT-IR analysis via lars predicted efficiently microbial loads of TVC.

Quest of Intelligent Research Tools for Rapid Evaluation of Fish Quality: FTIR Spectroscopy and Multispectral Imaging Versus Microbiological Analysis

The aim of the present study was to assess the microbiological quality of farmed sea bass (Dicentrarchus labrax) fillets stored under aerobic conditions and modified atmosphere packaging (MAP) (31% CO₂, 23% O₂, 46% Ν₂,) at 0, 4, 8, and 12 °C using Fourier transform infrared (FTIR) spectroscopy and multispectral imaging (MSI) in tandem with data analytics, taking into account the results of conventional microbiological analysis. Fish samples were subjected to microbiological analysis (total viable counts (TVC), Pseudomonas spp., H₂S producing bacteria, Brochothrix thermosphacta, lactic acid bacteria (LAB), Enterobacteriaceae, and yeasts) and sensory evaluation, together with FTIR and MSI spectral data acquisition. Pseudomonas spp. and H₂S-producing bacteria were enumerated at higher population levels compared to other microorganisms, regardless of storage temperature and packaging condition. The developed partial least squares regression (PLS-R) models based on the FTIR spectra of fish stored aerobically and under MAP exhibited satisfactory performance in the estimation of TVC, with coefficients of determination (R²) at 0.78 and 0.99, respectively. In contrast, the performances of PLS-R models based on MSI spectral data were less accurate, with R² values of 0.44 and 0.62 for fish samples stored aerobically and under MAP, respectively. FTIR spectroscopy is a promising tool to assess the microbiological quality of sea bass fillets stored in air and under MAP that could be effectively employed in the future as an alternative method to conventional microbiological analysis.

Authentication and Provenance of Walnut Combining Fourier Transform Mid-Infrared Spectroscopy with Machine Learning Algorithms

Different varieties and geographical origins of walnut usually lead to different nutritional values, contributing to a big difference in the final price. The conventional analytical techniques have some unavoidable limitations, e.g., chemical analysis is usually time-expensive and labor-intensive. Therefore, this work aims to apply Fourier transform mid-infrared spectroscopy coupled with machine learning algorithms for the rapid and accurate classification of walnut species that originated from ten varieties produced from four provinces. Three types of models were developed by using five machine learning classifiers to (1) differentiate four geographical origins; (2) identify varieties produced from the same origin; and (3) classify all 10 varieties from four origins. Prior to modeling, the wavelet transform algorithm was used to smooth and denoise the spectrum. The results showed that the identification of varieties under the same origin performed the best (i.e., accuracy = 100% for some origins), followed by the classification of four different origins (i.e., accuracy = 96.97%), while the discrimination of all 10 varieties is the least desirable (i.e., accuracy = 87.88%). Our results implicated that using the full spectral range of 700-4350 cm-1 is inferior to using the subsets of the optimal spectral variables for some classifiers. Additionally, it is demonstrated that back propagation neural network (BPNN) delivered the best model performance, while random forests (RF) produced the worst outcome. Hence, this work showed that the authentication and provenance of walnut can be realized effectively based on Fourier transform mid-infrared spectroscopy combined with machine learning algorithms.

Monitoring Thermal and Non-Thermal Treatments during Processing of Muscle Foods: A Comprehensive Review of Recent Technological Advances

Muscle food products play a vital role in human nutrition due to their sensory quality and high nutritional value. One well-known challenge of such products is the high perishability and limited shelf life unless suitable preservation or processing techniques are applied. Thermal processing is one of the well-established treatments that has been most commonly used in order to prepare food and ensure its safety. However, the application of inappropriate or severe thermal treatments may lead to undesirable changes in the sensory and nutritional quality of heat-processed products, and especially so for foods that are sensitive to thermal treatments, such as fish and meat and their products. In recent years, novel thermal treatments (e.g., ohmic heating, microwave) and non-thermal processing (e.g., high pressure, cold plasma) have emerged and proved to cause less damage to the quality of treated products than do conventional techniques. Several traditional assessment approaches have been extensively applied in order to evaluate and monitor changes in quality resulting from the use of thermal and non-thermal processing methods. Recent advances, nonetheless, have shown tremendous potential of various emerging analytical methods. Among these, spectroscopic techniques have received considerable attention due to many favorable features compared to conventional analysis methods. This review paper will provide an updated overview of both processing (thermal and non-thermal) and analytical techniques (traditional methods and spectroscopic ones). The opportunities and limitations will be discussed and possible directions for future research studies and applications will be suggested.