Partial least squares density modeling (PLS-DM) – A new class-modeling strategy applied to the authentication of olives in brine by near-infrared spectroscopy

A digital image smartphone-based approach to Slovak Tokaj wine authentication chemometric assisted

The authentication of Slovak wines in comparison to other similar wines from various geographical regions, namely Hungary, France, Austria, and Ukraine, was conducted using the OC-PLS, DD-SIMCA, and PLS-DM models, all of them operating in rigorous way. The study involved 63 samples, of which 41 originated from Slovakia, covering diverse wine types such as varietal wines, cuvée selections (different "putňový"), and essence. To capture digital images under controlled conditions, a custom-made cardboard box with white inner surfaces was devised and equipped with a smartphone. During the training phase, sensitivities of 96%, 100%, and 96% were attained for OC-PLS, DD-SIMCA, and PLS-DM, respectively. In the subsequent stages of validation and testing for DD-SIMCA and PLS-DM, the proposed methods displayed optimal efficiency, achieving both sensitivity and specificity rates of 100%. However, such results were not achieved in the case of OC-PLS, which exhibited efficiency levels of 90% in validation and 80% in testing.

In-house validation of a visible and near infrared spectroscopy non-targeted method to support panel test of virgin olive oils

In this study, an in-house validation of Visible and Near Infrared Spectroscopy was performed to distinguish between extra virgin olive oil (EVOO) and virgin olive oil (VOO). A total of 161 samples of olive oil of three different categories (EVOO, VOO and lampante (LOO)) were analysed by transflectance using a monochromator instrument. One-class models were initially developed using Partial Least Squares (PLS) Density Modelling to characterize EVOO and VOO category. Once the LOO samples were discriminated, linear and non-linear discriminant models were built to classify EVOO and VOO. Different data pre-treatments and variable selection algorithms were evaluated to establish the best models in terms of Correct Classification Rate (CCR). The best model, obtained after variable selection using PLS Discriminant Analysis, yielded CCR values of 82.35 % for EVOO and 66.67 % for VOO in external validation. These results confirmed that VIS + NIRS technology may be used to provide rapid, non-destructive preliminary screening of olive oil samples for categorization; suspect samples may then be analysed by official analytical methods.

Optimizing the soft independent modeling of class analogy (SIMCA) using statistical prediction regions

The ultimate goal of a one-class classifier like the "rigorous" soft independent modeling of class analogy (SIMCA) is to predict with a certain confidence probability, the conformity of future objects with a given reference class. However, the SIMCA model, as currently implemented often suffers from an undercoverage problem, meaning that its observed sensitivity often falls far below the desired theoretical confidence probability, hence undermining its intended use as a predictive tool. To overcome the issue, the most reported strategy in the literature, involves incrementing the nominal confidence probability until the desired sensitivity is obtained in cross-validation. This article proposes a statistical prediction interval-based strategy as an alternative strategy to properly overcome this undercoverage issue. The strategy uses the concept of predictive distributions sensu stricto to construct statistical prediction regions for the metrics. Firstly, a procedure based on goodness-of-fit criteria is used to select the best-fitting family of probability models for each metric or its monotonic transformation, among several plausible candidate families of right-skewed probability distributions for positive random variables, including the gamma and the lognormal families. Secondly, assuming the best-fitting distribution, a generalized linear model is fitted to each metric data using the Bayesian method. This method enables to conveniently estimate uncertainties about the parameters of the selected distribution. Propagating these uncertainties to the best-fitting probability model of the metric enables to derive its so-called posterior predictive distribution, which is then used to set its critical limit. Overall, the evaluation of the proposed approach on a diversity of real datasets shows that it yields unbiased and more accurate sensitivities than existing methods which are not based on predictive densities. It can even yield better specificities than the strategy that attempts to improve sensitivities of existing methods by "optimizing" the type 1 error, especially in low sample sizes' contexts.

Authentication and Chemometric Discrimination of Six Greek PDO Table Olive Varieties through Morphological Characteristics of Their Stones

Table olives, the number one consumed fermented food in Europe, are widely consumed as they contain many valuable ingredients for health. It is also a food which may be the subject of adulteration, as many different olive varieties with different geographical origin, exist all over the word. In the present study, the image analysis of stones of six main Greek protected designation of origin (PDO) table olive varieties was performed for the control of their authentication and discrimination, with cv. Prasines Chalkidikis, cv. Kalamata Olive, cv. Konservolia Stylidas, cv. Konservolia Amfissis, cv. Throuba Thassos and cv. Throuba Chios being the studied olive varieties. Orthogonal partial least square discriminant analysis (OPLS-DA) was used for discrimination and classification of the six Greek table olive varieties. With a 98.33% of varietal discrimination, the OPLS-DA model proved to be an efficient tool to authentify table olive varieties from their morphological characteristics.

Emerging nondestructive approaches for meat quality and safety evaluation-A review

Meat is one of the most consumed agro-products because it contains proteins, minerals, and essential vitamins, all of which play critical roles in the human diet and health. Meat is a perishable food product because of its high moisture content, and as such there are concerns about its quality, stability, and safety. There are two widely used methods for monitoring meat quality attributes: subjective sensory evaluation and chemical/instrumentation tests. However, these methods are labor-intensive, time-consuming, and destructive. To overcome the shortfalls of these conventional approaches, several researchers have developed fast and nondestructive techniques. Recently, electronic nose (e-nose), computer vision (CV), spectroscopy, hyperspectral imaging (HSI), and multispectral imaging (MSI) technologies have been explored as nondestructive methods in meat quality and safety evaluation. However, most of the studies on the application of these novel technologies are still in the preliminary stages and are carried out in isolation, often without comprehensive information on the most suitable approach. This lack of cohesive information on the strength and shortcomings of each technique could impact their application and commercialization for the detection of important meat attributes such as pH, marbling, or microbial spoilage. Here, we provide a comprehensive review of recent nondestructive technologies (e-nose, CV, spectroscopy, HSI, and MSI), as well as their applications and limitations in the detection and evaluation of meat quality and safety issues, such as contamination, adulteration, and quality classification. A discussion is also included on the challenges and future outlooks of the respective technologies and their various applications.

Chemometric Strategies for Spectroscopy-Based Food Authentication

In the last decades, spectroscopic techniques have played an increasingly crucial role in analytical chemistry, due to the numerous advantages they offer. Several of these techniques (e.g., Near-InfraRed—NIR—or Fourier Transform InfraRed—FT-IR—spectroscopy) are considered particularly valuable because, by means of suitable equipment, they enable a fast and non-destructive sample characterization. This aspect, together with the possibility of easily developing devices for on- and in-line applications, has recently favored the diffusion of such approaches especially in the context of foodstuff quality control. Nevertheless, the complex nature of the signal yielded by spectroscopy instrumentation (regardless of the spectral range investigated) inevitably calls for the use of multivariate chemometric strategies for its accurate assessment and interpretation. This review aims at providing a comprehensive overview of some of the chemometric tools most commonly exploited for spectroscopy-based foodstuff analysis and authentication. More in detail, three different scenarios will be surveyed here: data exploration, calibration and classification. The main methodologies suited to addressing each one of these different tasks will be outlined and examples illustrating their use will be provided alongside their description.

Chemometric tools for food fraud detection: The role of target class in non-targeted analysis

The chemometric issues related to the application of non-targeted analysis for the detection of food frauds were analyzed employing discriminant analysis and a one-class classifier. The similarities and differences between the two methods were investigated. The results of classification are characterized by a set of indices called figures of merit. They comprehensively characterized the quality and reliability of classification. The principle is illustrated using an actual example of Oregano herbs adulteration. The informative region 9000-4000 cm^-1 of near-Infrared spectroscopy is used as analytical means. The results of the application of each method for Oregano data collection are presented. It is shown that the discriminant method is only partially appropriate for solving the authentication problem. One class classifier is a powerful and devoted for non-targeted analysis. The step by step analysis introduced in the paper can also be successfully utilized in apply for revealing of forgeries of various food products.

Random forest as one-class classifier and infrared spectroscopy for food adulteration detection

This paper proposes the use of random forest for adulteration detection purposes, combining the random forest algorithm with the artificial generation of outliers from the authentic samples. This proposal was applied in two food adulteration studies: evening primrose oils using ATR-FTIR spectroscopy and ground nutmeg using NIR diffuse reflectance spectroscopy. The primrose oil was adulterated with soybean, corn and sunflower oils, and the model was validated using these adulterated oils and other different oils, such as rosehip and andiroba, in pure and adulterated forms. The ground nutmeg was adulterated with cumin, commercial monosodium glutamate, soil, roasted coffee husks and wood sawdust. For the primrose oil, the proposed method presented superior performance than PLS-DA and similar performance to SIMCA and for the ground nutmeg, the random forest was superior to PLS-DA and SIMCA. Also, in both applications using the random forest, no sample was excluded from the external validation set.

Detection of adulterants in grape nectars by attenuated total reflectance Fourier-transform mid-infrared spectroscopy and multivariate classification strategies

There is no any doubt about the importance of food fraud control, as it has implications in food safety and in consumer health. Focusing on fruit beverages, some types of adulterations have been detected more frequently, such as substitution with less expensive fruits. A methodology based on attenuated total reflectance Fourier-transform mid-infrared spectroscopy (ATR-FTIR) and multivariate classification was applied to detect whether grape nectars were adulterated by substitution with apple juice or cashew juice. A total of 126 samples were obtained and analyzed. Two strategies were proposed: one-class and multiclass approaches. Soft independent modeling of class analogy (SIMCA), partial least squares discriminant analysis (PLS-DA) and partial least squares density modeling (PLS-DM) were used to build the models. Among them, PLS-DA presented the best performance with a sensitivity and specificity of nearly 100%. The multiclass strategy was preferred if the adulterants to be studied are known because it provides additional information.

Class-modelling in food analytical chemistry: Development, sampling, optimisation and validation issues - A tutorial

Qualitative data modelling is a fundamental branch of pattern recognition, with many applications in analytical chemistry, and embraces two main families: discriminant and class-modelling methods. The first strategy is appropriate when at least two classes are meaningfully defined in the problem under study, while the second strategy is the right choice when the focus is on a single class. For this reason, class-modelling methods are also referred to as one-class classifiers. Although, in the food analytical field, most of the issues would be properly addressed by class-modelling strategies, the use of such techniques is rather limited and, in many cases, discriminant methods are forcedly used for one-class problems, introducing a bias in the outcomes. Key aspects related to the development, optimisation and validation of suitable class models for the characterisation of food products are critically analysed and discussed.

A NIR spectroscopy-based efficient approach to detect fraudulent additions within mixtures of dried porcini mushrooms

Boletus edulis and allied species (BEAS), known as "porcini mushrooms", represent almost the totality of wild mushrooms placed on the Italian market, both fresh and dehydrated. Furthermore, considerable amounts of these dried fungi are imported from China. The presence of Tylopilus spp. and other extraneous species (i.e., species edible but not belonging to BEAS) within dried porcini mushrooms - mainly from those imported from China and sold in Italy - may represent an evaluable problem from a commercial point of view. The purpose of the present study is to evaluate near-infrared spectroscopy (NIRS) as a rapid and effective alternative to classical methods for identifying extraneous species within dried porcini batches and detecting related commercial frauds. To this goal, 80 dried fungi including BEAS, Tylopilus spp., and Boletus violaceofuscus were analysed by NIRS. For each sample, 3 different parts of the pileus (pileipellis, flesh and hymenium) were analysed and a low-level strategy for data fusion, consisting of combining the signals obtained by the different parts before data processing, was applied. Then, NIR spectra were used to develop reliable and efficient class-models using a novel method, partial least squares density modelling (PLS-DM), and the two most commonly used class-modelling techniques, UNEQ and SIMCA. The results showed that NIR spectroscopy coupled with chemometric class-modelling technique can be suggested as an effective analytical strategy to check the authenticity of dried BEAS mushrooms.

Optimization Control of the Color-Coating Production Process for Model Uncertainty

Optimized control of the color-coating production process (CCPP) aims at reducing production costs and improving economic efficiency while meeting quality requirements. However, because optimization control of the CCPP is hampered by model uncertainty, a strategy that considers model uncertainty is proposed. Previous work has introduced a mechanistic model of CCPP based on process analysis to simulate the actual production process and generate process data. The partial least squares method is then applied to develop predictive models of film thickness and economic efficiency. To manage the model uncertainty, the robust optimization approach is introduced to improve the feasibility of the optimized solution. Iterative learning control is then utilized to further refine the model uncertainty. The constrained film thickness is transformed into one of the tracked targets to overcome the drawback that traditional iterative learning control cannot address constraints. The goal setting of economic efficiency is updated continuously according to the film thickness setting until this reaches its desired value. Finally, fuzzy parameter adjustment is adopted to ensure that the economic efficiency and film thickness converge rapidly to their optimized values under the constraint conditions. The effectiveness of the proposed optimization control strategy is validated by simulation results.

One-class classification based authentication of peanut oils by fatty acid profiles

In this study, the authenticity identification model was built by the one-class partial least squares (OCPLS) classifier for peanut oils, which could effectively detect adulterated oils at the adulteration level of more than 4%.