Soft Independent Modeling By Class Analogy Research Articles

Development of a new and rapid FTIR method using chemometric modeling techniques for the determination of lavandin adulteration in lavender essential oil

Lavender essential oil (LT) is subject to adulteration by malicious suppliers and is often sold mixed with the cheaper lavandin essential oil (LD). In this study, it is aimed to develop a new and rapid spectroscopic method by using chemometric modeling techniques for the qualitative and quantitative determination of LD adulteration in LT. For the discrimination of LD in LT, 36 principal component analysis (PCA) models were created and the model with the best statistical results was evaluated by the number of PCs in the explained variance and eigen value. Soft independent modeling by class analogy (SIMCA) was also employed to demonstrate the discrimination capabilities of the best PCA model. For the quantification of LD in LT, 36 partial least squares (PLS) regression models employing the spectral data were developed. To find out the best PLS model, the root mean square error of prediction (RMSEP), and root mean square error of cross-validation (RMSECV) were mainly used. It was found that the best models for discrimination and quantitative analysis were those using normal spectra in the region 2200–400 cm−1 (NDSR). The results obtained showed that LD adulteration in LT could be successfully detected over a wide concentration range, thanks to the combined use of FTIR spectroscopy and chemometry without using any toxic chemicals or pretreatments.

Green Chemometric-Assisted Characterization of Common and Black Varieties of Celery.

Celery (Apium graveolens L., var. Dulce), is a biennial herbaceous plant belonging to the Apiaceae family, cultivated in humid soils in the Mediterranean basin, in Central-Southern Europe, and in Asia. Despite its wide diffusion and although it is well-known that cultivar/origin strongly influences plant composition, only a few studies have been carried out on the different types of celery. The present work aims to investigate four different Italian types of celery (two common, Elne and Magnum celery, and two black, Torricella Peligna Black and Trevi Black celery), and to test, whether the combination of FT-IR spectroscopy and chemometrics allows their ecotype discrimination. The peculiarity of this study lies in the fact that all the analyzed celeries were grown in the same experimental field under the same soil and climate conditions. Consequently, the differences captured by the FT-IR-based tool are mainly imputable to the different ecotypes. In order to achieve this goal, FT-IR profiles were handled by two diverse classifiers: sequential preprocessing through ORThogonalization (SPORT) and soft independent modeling by class analogy (SIMCA). Eventually, the highest classification rate (90%, on an external set of 100 samples) has been achieved by SPORT.

Classification and authentication of spices and aromatic herbs by means of HPLC-UV and chemometrics

Recent increase in the adulteration of spices and aromatic herbs in food industry constitutes a problem that requires exhaustive quality control. As every spice has a different composition with characteristic biomarkers, chromatographic profiles are especially valuable to authenticate these products. Thus, in this work a new high performance liquid chromatography (HPLC) method with UV–vis detection was developed for the characterization, identification and authentication of cinnamon, oregano, thyme, sesame, bay leaf, clove, cumin, and vanilla. Chromatographic separation was optimized based on the separation of six characteristic biomarkers (sesamol, eugenol, thymol, carvacrol, salicylaldehyde and vainillin) and was performed using a C18 reversed-phase column under a 35 min gradient elution based on 0.1% (v/v) formic aqueous solution and methanol by means of UV–Vis detection at 280 nm. 87 samples, purchased in local supermarkets, were analyzed and the obtained profiles were processed by chemometric techniques. First, data treatment was evaluated by principal component analysis (PCA); next soft independent modelling by class analogy (SIMCA) and partial least squares discriminant analysis (PLS-DA) were carried out in order to verify if classification according to their biomarkers was possible. The study concluded that PLS-DA (0.14–0.75% global error) classifies better the types of spice or aromatic herb than SIMCA (0.82–3.67% global error).

Metabolomics as a Potential Chemotaxonomical Tool: Application on the Selected Euphorbia Species Growing Wild in Serbia.

Chemotaxonomy presents various challenges that need to be overcome in order to obtain valid and reliable results. Individual genetic and environmental variations can give a false picture and lead to wrong conclusions. Applying a holistic approach, based on multivariate data analysis, these challenges can be overcome. Thus, a metabolomics approach has to be optimized depending on the subject of research. We used 1H NMR-based metabolomics as a potential chemotaxonomic tool on the selected Euphorbia species growing wild in Serbia. Principal components analysis (PCA), soft independent modeling by class analogy (SIMCA) and Orthogonal Projections to Latent Structures Discriminant Analysis (OPLS-DA) were used to analyze obtained NMR data in order to reveal chemotaxonomic biomarkers. The standard protocol for plant metabolomics was optimized aiming to extract more specific metabolites, which are characteristic for the Euphorbia genus. The obtained models were validated, which revealed that variables unique for each species were associated with certain classes of molecules according to literature data. In E. salicifolia, acacetin-7-O-glycoside (not found before in the species) was detected, and the structure of the aglycone part was solved based on 2D NMR data. In the presented paper, we have shown that metabolomics can be successfully used in Euphorbia chemotaxonomy.

Saffron Characterization by a Multidisciplinary Approach

Saffron is a spice obtained from the drying process of the stigmas of the flower Crocus sativus Linnaeus. It is well known that the organoleptic characteristics of this spice are closely linked to the production area and harvesting year. The present work aims to evaluate whether saffron samples produced in different years and origins present sensibly different crocin profiles. To achieve this goal, 120 saffron samples were harvested between 2016 and 2020 in four different Italian areas. The crocins were analysed, identified, and quantified by high-performance liquid chromatography-electrospray-tandem mass spectrometry (HPLC-ESI-MS/MS) in multiple reaction monitoring mode (MRM). Subsequently, ANOVA-simultaneous component analysis (ASCA) was used to evaluate whether the origin and annuity significantly affected the composition of the crocins. ASCA confirmed the relevance of these effects. Eventually, soft independent modelling by class analogy (SIMCA) models were created for each of the four different origins. Mixtures of saffron from different areas were also prepared to test the robustness of the models. SIMCA provided satisfying results; in fact, models provided 100% sensitivity for three origins (Cascia, Sardinia, and Città della Pieve) on the external test set (48 samples) and 88% (sensitivity on the external test set) for the Spoleto class.

Characterization and Authentication of "Ricotta" Whey Cheeses through GC-FID Analysis of Fatty Acid Profile and Chemometrics.

The fatty acid (FA) profiles of 240 samples of ricotta whey cheese made from sheep, goat, cow, or water buffalo milk were analyzed by gas-chromatography (GC). Then, sequential preprocessing through orthogonalization (SPORT) was used in order to classify samples according to the nature of the milk they were made from. This strategy achieved excellent results, correctly classifying 77 (out of 80) validation samples. Eventually, since 36 (over 114) sheep ricotta whey cheeses were PDO products, a second classification problem, finalizing the discrimination of PDO and Non-PDO dairies, was faced. In this case, two classifiers were used, SPORT and soft independent modelling by class analogy (SIMCA). Both approaches provided more than satisfying results; in fact, SPORT properly assigned 63 (of 65) test samples, whereas the SIMCA model accepted 14 PDO individuals over 15 (93.3% sensitivity) and correctly rejected all the other samples (100.0% specificity). In conclusion, all the tested approaches resulted as suitable for the two fixed purposes. Eventually, variable importance in projection (VIP) analysis was used to understand which FAs characterize the different categories of ricotta. Among the 22 analyzed compounds, about 10 are considered the most relevant for the solution of the investigated problems.

Authentication of typical Italian bell pepper spices by ICP-OES multi-elemental analysis combined with SIMCA class modelling

Ground bell pepper is a spice used worldwide. The cultivation area has a significant influence on the chemical composition of this food product; consequently, its origin is closely linked to the organoleptic characteristics. This study is focused on the analysis of the mineral composition of two high-valued Italian ground bell peppers: the Senise PGI and the Altino varieties. Fifty samples of bell pepper spices from these two origins and 20 commercial samples were analysed by Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES), with the aim of quantifying thirteen elements: K, B, Ba, Ca, Cu, Fe, Mg, Mn, Na, Ni, P, Sr and Zn. Eventually, the outcome of the ICP-OES analysis was handled by Soft Independent Modelling by Class Analogy (SIMCA) to characterize the two different products. Two individual SIMCA models were developed, one for each origin; both models achieved 100% of sensitivity and specificity in prediction.

Identification of adulterants (sunflower oil and soybean oil) in grapeseed oil (Vitis vinifera L.) and chia oil (Salvia hispanica L.) by FT-MIR spectroscopy and chemometric

Grapeseed and chia oils are important for their fatty acids profile. Both oils are considered gourmet and due their high retail prices they are susceptible to adulteration. FT-MIR was used with chemometric analysis to identify and quantify the adulterants (sunflower and soybean oils) in grapeseed and chia oils. A ternary mixture system was designed for each oil using as adulterants sunflower and soybean oils (2-50%, w/w). Soft independent modelling by class analogy (SIMCA) discriminated adulterated samples and pure samples (100% of recognition and rejection). Partial least squares (PLS) showed the best calibration result (R2c: ≥ 0.95; SEC: 2.22-2.87) and validation (R2v: ≥ 0.99; SEP: 2.25-3.13). The models could be applied to identify and quantify two adulterants: sunflower and soybean oils in grapeseed and chia oils

Forensic Analysis of Textile Synthetic Fibers Using a FT-IR Spectroscopy Approach.

Synthetic fibers are one of the most valuable trace lines of evidence that can be found in crime scenes. When textile fibers are analyzed properly, they can help in finding a linkage between suspect, victim, and the scene of the crime. Various analytical techniques are used in the examination of samples to determine relationships between different fabric fragments. In this exploratory study, multivariate statistical methods were investigated in combination with machine learning classification models as a method for classifying 138 synthetic textile fibers using Fourier transform infrared spectroscopy, FT-IR. The data were first subjected to preprocessing techniques including the Savitzky–Golay first derivative method and Standard Normal Variate (SNV) method to smooth the spectra and minimize the scattering effects. Principal Component Analysis (PCA) was built to observe unique patterns and to cluster the samples. The classification model in this study, Soft Independent Modeling by Class Analogy (SIMCA), showed correct classification and separation distances between the analyzed synthetic fiber types. At a significance level of 5%, 97.1% of test samples were correctly classified.

Development of methods for quantification of gelatin in vegetable capsule shell and for classification of capsule shell as from vegetable or animal sources by chemometric analysis of FTIR spectroscopic data

Capsule shell from animal source (bovine or porcine gelatin) is a problem for the follower of different religions and vegetarian. In that case, vegetable capsule shell could be a solution. In this research, we proposed a simple and cost-effective technique for detection of gelatin in vegetable capsule shell and for classification of capsule shell by source, based on Chemometric techniques with FTIR spectroscopic data. Partial Least-Square Regression (PLSR) models were developed and their efficiencies were assessed with spectroscopic data of range of 4000-700 cm-1. PLSR shows very good prediction efficiency (R2= 98%) for both vegetable capsule shell and gelatin. In addition, Soft Independent Modeling by Class Analogy (SIMCA) classification method were developed and assessed with spectral data of capsule shells from vegetable and animal sources. Results prove that FTIR spectroscopy in combination with chemometric method can be used for the classification of capsule shell by source and quantification of gelatin in vegetable capsule shell to ensure their authenticity.
 Bangladesh J. Sci. Ind. Res. 57(2), 91-98, 2022

Untargeted selected ion flow tube mass spectrometry headspace analysis: High-throughput differentiation of virgin and recycled polyethylene pellets.

Recycled plastics are increasingly used for packaging of fast-moving consumer goods (FMCG). Compared with packaging made from virgin polymers, there is greater risk of taints entering products due to prior use of the polymers and incomplete cleaning. Increased quality assurance testing of polymer feedstock is required for recycled packaging. Selected ion flow tube mass spectrometry (SIFT-MS) analysis coupled with multivariate statistical data processing can provide high-throughput untargeted screening of recycled polymers at low cost per sample. SIFT-MS is a direct-injection MS technique that provides high-throughput automated headspace analysis of polymer samples when coupled with a syringe-injection autosampler (12 incubated samples per hour). Full-scan SIFT-MS data were processed using multivariate statistical analysis (specifically, the soft independent modeling by class analogy (SIMCA) algorithm). SIFT-MS full-scan data were acquired for ten replicates each of ten recycled and four virgin high-density polyethylene (HDPE) pellet products from multiple manufacturers. The samples varied approximately 20-fold in terms of total volatile residue, while showing very high repeatability across replicates. SIFT-MS scan data were dominated by aliphatic and monoterpene hydrocarbon residues, and - to a lesser extent - alcohols. Application of the SIMCA algorithm to the data resulted in successful classification by both individual samples and manufacturers. Automated, untargeted SIFT-MS analysis coupled with multivariate statistical data analysis has the potential to provide rapid, effective screening of recycled polymer products, which would provide increased quality assurance of recycled polymers used for FMCG.

Detection of coconut oil in ghee using ATR-FTIR and chemometrics

Ghee, the most valuable fat is sold at a premium price over other fats and oils. Unscrupulous sellers take this advantage by mixing it with less expensive fats or oils. The addition of coconut oil (2, 4, 6, 8, 10 and 15%) in ghee was detected using ATR-FTIR and chemometrics. The spectra of pure ghee, coconut oil and adulterated samples (total 240 samples) were analyzed in the wavenumber region of 4000–500 cm−1. Principal component analysis (PCA) showed the distinct grouping of pure ghee and adulterated samples in the selected wavenumber range (1170-1141 and 1117–1100 cm−1). Soft independent modelling by class analogy (SIMCA) was able to categorize the pure ghee and coconut oil samples of the validation set with a classification efficiency of 100% using both Partial Least Squares Regression (PLS) and Principal components regression (PCR) models constructed on the training set. R2 values of greater than 0.99 between the actual and predicted values of adulterated ghee with coconut oil were observed in both the calibration and validation sets for the developed PLS and PCR models. The study revealed that coconut oil adulterated samples can be detected even at a lower concentration of 2% in ghee.

Electronic tongue based on a single impedimetric sensor and complex numbers-supervised pattern recognition

This work investigated the application of supervised pattern recognition of complex number data obtained by electrochemical impedance spectroscopy (EIS) in order to obtain qualitative/discriminatory electroanalytical methods based on electronic tongues/noses. As a case study, electrochemical impedance spectra of aqueous extracts of coffee samples were obtained with a single label-free sensor and submitted to complex numbers-supervised pattern recognition to detect the adulteration of ground roasted coffee by the addition of coffee husks and sticks. Partial least-squares discriminant analysis (PLS-DA) and soft independent modeling by class analogy (SIMCA), two of the most employed multivariate classification techniques, were evaluated to classify coffee samples as adulterated or unadulterated by the direct analysis of complex impedance spectra. They are techniques whose mathematical logic can be easily employed in the treatment of complex numbers with very little or even no modification, by using the Hermitian transpose instead of the common nonconjugate transpose. The predictive ability of complex numbers–supervised pattern recognition on the identification of adulterated ground roasted coffees was compared to that obtained with impedance data presented as real numbers – real (Z’), imaginary (Z”), absolute impedance (|Z|), and phase angle (ϕ). A discussion about the most suitable classification algorithm for the type of data employed was held. The highest predictive ability was obtained by applying complex numbers-PLS-DA, with 100 % and 96 % of the adulterated and unadulterated test samples correctly identified, respectively, with no need for reducing complex spectra to real numbers.

Untargeted Metabolomics in Forensic Toxicology: A New Approach for the Detection of Fentanyl Intake in Urine Samples.

The misuse of fentanyl, and novel synthetic opioids (NSO) in general, has become a public health emergency, especially in the United States. The detection of NSO is often challenged by the limited diagnostic time frame allowed by urine sampling and the wide range of chemically modified analogues, continuously introduced to the recreational drug market. In this study, an untargeted metabolomics approach was developed to obtain a comprehensive “fingerprint” of any anomalous and specific metabolic pattern potentially related to fentanyl exposure. In recent years, in vitro models of drug metabolism have emerged as important tools to overcome the limited access to positive urine samples and uncertainties related to the substances actually taken, the possible combined drug intake, and the ingested dose. In this study, an in vivo experiment was designed by incubating HepG2 cell lines with either fentanyl or common drugs of abuse, creating a cohort of 96 samples. These samples, together with 81 urine samples including negative controls and positive samples obtained from recent users of either fentanyl or “traditional” drugs, were subjected to untargeted analysis using both UHPLC reverse phase and HILIC chromatography combined with QTOF mass spectrometry. Data independent acquisition was performed by SWATH in order to obtain a comprehensive profile of the urinary metabolome. After extensive processing, the resulting datasets were initially subjected to unsupervised exploration by principal component analysis (PCA), yielding clear separation of the fentanyl positive samples with respect to both controls and samples positive to other drugs. The urine datasets were then systematically investigated by supervised classification models based on soft independent modeling by class analogy (SIMCA) algorithms, with the end goal of identifying fentanyl users. A final single-class SIMCA model based on an RP dataset and five PCs yielded 96% sensitivity and 74% specificity. The distinguishable metabolic patterns produced by fentanyl in comparison to other opioids opens up new perspectives in the interpretation of the biological activity of fentanyl.

MIR spectroscopy versus MALDI-ToF-MS for authenticity control of honeys from different botanical origins based on soft independent modelling by class analogy (SIMCA) – A clash of techniques?

In this study, highly reproducible MIR spectroscopy and highly sensitive MALDI-ToF-MS data were directly compared for the metabolomic profiling of monofloral and multifloral honey samples from three different botanical origins canola, acacia, and honeydew. Subsequently, three different classification models were applied to the data of both techniques, PCA-LDA, PCA- kNN, and soft independent modelling by class analogy (SIMCA) as class modelling technique. All monofloral external test set samples were classified correctly by PCA-LDA and SIMCA with both data sets, while multifloral test set samples could only be identified as outliers by the SIMCA technique, which is a crucial aspect in the authenticity control of honey. The comparison of the two used analytical techniques resulted in better overall classification results for the monofloral external test set samples with the MIR spectroscopic data. Additionally, clearly more multifloral external samples were identified as outliers by MIR spectroscopy (91.3%) as compared to MALDI-ToF-MS (78.3%). The results indicate that the high reproducibility of the used MIR technique leads to a generally better ability of separating monofloral honeys and in particular, identifying multifloral honeys. This demonstrates that benchtop-based techniques may operate on an eye-level with high-end laboratory-based equipment, when paired with an optimal data analysis strategy.

Feasibility study on the use of elemental profiles to authenticate aromatic rice: the case of Basmati and Thai rice

Among the thousands of existing rice varieties, aromatic rice has increasingly attracted consumer’s preference in recent years. Within aromatic rice, Basmati, cultivated in some regions in Pakistan and India, is highly demanded. Other aromatic rice, cultivated in specific regions, for instance in Thailand (commonly referred to as Jasmine Thai rice), are also highly appreciated by consumers. In this work, the elemental profiles of commercially available rice samples (17 Basmati, 11 Thai, and 7 Long Grain rice) were determined by energy-dispersive X-ray fluorescence (ED-XRF) spectroscopy. The mass fractions of P, Cl, S, K, Fe, Cu, and Zn were significantly different (95% confidence interval) between Basmati and Thai rice and between Thai and Long Grain rice; only Cl, S, and Zn were significantly different between Basmati and Long Grain rice. Multivariate evaluation of the results combining soft independent modelling by class analogy (SIMCA) and partial least square discriminant analysis (PLS-DA) allowed the correct classification (true positives) of 94.1, 85.6, and 100% of the Basmati, Long Grain, and Thai rice, respectively. The specificity (true negatives) of Basmati, Long Grain, and Thai was 94.4, 82.1, and 100%, respectively.

Trace elements in dried blood spots as potential discriminating features for metabolic disorder diagnosis in newborns

Trace elements in dried blood spots (DBSs) from newborns were determined by laser ablation coupled with inductively coupled plasma mass spectrometry, and data were subjected to chemometric evaluation in an attempt to classify healthy newborns and newborns suffering from metabolic disorders. Unsupervised [principal component analysis (PCA) and cluster analysis (CA)] and supervised [linear discriminant analysis (LDA) and soft independent modeling by class analogy (SIMCA)] pattern recognition techniques were used as classification techniques. PCA and CA have shown a clear tendency to form two groups (healthy newborns and newborns suffering from metabolic disorders). LDA and SIMCA have predicted that 90.5% and 83.9% of originally grouped healthy newborn cases were correctly classified by LDA and SIMCA, respectively. In addition, these percentages were 97.6% (LDA) and 80.6% (SIMCA) for DBSs from newborns suffering from metabolic disorders. However, SIMCA has only detected one misclassified DBS from the healthy group, and the lower percentage is attributed to four DBSs from the healthy newborn group and five DBSs from newborns with disorders that were found as belonging to both categories (healthy newborns and newborns with disorders) in the training set. LDA also gave a percentage of grouped maple syrup urine disease (MSUD) cases correctly classified of 100%, although the percentage fells to 66.7% when classifying phenylketonuria (PKU) cases. Finally, essential elements such as Fe, K, Rb, and Zn were found to be matched (correlated) with the concentration of amino acids such as phenylalanine, valine, and leucine, biomarkers linked with MSUD and PKU diseases.

Cleaner and faster method to detect adulteration in cassava starch using Raman spectroscopy and one-class support vector machine

Due to food adulteration concerns, analytical assays are routinely performed in labs to evaluate and ensure food quality control. However, classical analytical methods used to acquire reliable results are lengthy and costly. Therefore, we aim to propose a new approach to detect adulterants in cassava starch in a clean, green, cheap, and quick way. Raman spectroscopy meets all these requirements and presents great potential to perform such routine analyses. Data treatment is also an important step in authentication problems, and we propose the use of one-class models to do so. One-class support vector machine (OC-SVM) and soft independent modelling by class analogy (SIMCA) were the two approaches to one-class classifiers assessed in this study. Cassava starch samples were modified in the lab with adulteration ranging from 0.5 to 50%, with adulterants such as wheat flour, sodium bicarbonate, and others. The two chemometric models were statistically compared and OC-SVM was found to outperform SIMCA, reaching higher values of sensitivity (87.1%), specificity (86.8%), and accuracy (86.9%) in the prediction of known data samples. This better performance also resulted in the possibility of detecting adulterations over 2% by OC-SVM, compared to only 5% by SIMCA.

Attenuated Total Reflection Fourier Transform Infrared Spectroscopy combined with chemometric modelling for the classification of clinically relevant Enterococci.

Attenuated Total Reflection Fourier Transform Infrared (ATR-FT-IR) Spectroscopy and chemometric modelling, including soft independent modelling by class analogy (SIMCA), partial least squares discriminant analysis (PLS-DA) and support vector machine (SVM), were applied to attempt to discriminate 60 clinical isolates of Enterococcus faecium and Enterococcus faecalis and hence evaluate the performance of the spectroscopic approach in identifying enterococci infections. The bacterial samples were identified by polymerize chain reaction (PCR) amplification and their ATR-FT-IR spectra acquired. Spectra were processed to the second derivative using the Savitzky-Golay algorithm and normalized using extended multiplicative signal correction employing the UnscramblerX (CAMO, Norway) software package. Multivariate classification models and their performance were evaluated using Cohen's Kappa coefficient. Principal component analysis (PCA) score plots showed separate clusters of spectra related to membership to E. faecium and E. faecalis, with this explained by bands assigned to PO2 (1230cm-1 ), P-O-C (1114cm-1 ), monosubstituted alkene (997, 987cm-1 ) and C-O (1070, 1055, 1036cm-1 ) corresponding to teichoic acids, polysaccharides and peptidoglycan from the cell wall in PCA and PLS-DA loading plots. The best classification model for E. faecium and E. faecalis is SVM, indicating via highest Kappa score. The classification coefficient between SIMCA, PLS-DA, SVM and PCR as reference method were 0·59, 0·9 and 1, respectively, shown as the Kappa scores. The main spectral differences observed between the two clinically relevant enterococci species were associated with changes in the teichoic acid content of cell walls. With regard to the binary classification method, SVM was found to be the best performing classification model, providing the highest correlation with the PCR results. The study shows that ATR-FT-IR spectroscopy in combination with chemometric modelling can be applied for the phenotypic identification and discrimination of clinically relevant and similar enterococcal species.

Authentication of the Geographical Origin of “Vallerano” Chestnut by Near Infrared Spectroscopy Coupled with Chemometrics

Promoting and protecting the market of typical national territory’s products are fundamental from a commodity point of view. Often, high-added value foods are subjected to fraud. Chestnuts have extraordinary nutritional and organoleptic qualities and Italy is one of the biggest producers of this product. The purpose of the present study is to develop an analytical method suitable to authenticate the chestnut of Vallerano a PDO agro-food produced in Central-Italy. A total of 441 chestnuts were analyzed (323 PDO and 118 harvested in other Italian territories) by near infrared spectroscopy (NIR) and then classified using two different approaches: a discriminant one, partial least squares-discriminant analysis (PLS-DA), and a class-modelling one, soft independent modelling by class analogy (SIMCA). Both strategies led to very high prediction capability in external validation on a test set (classification accuracy in one case, and sensitivity and specificity in the other). Eventually, both the proposed approaches resulted suitable for a rapid and non-destructive authentication of this valuable product. In particular, the combination of NIR (collected on the hilum) and PLS-DA provided the best results, reaching 97.0% of total correct classification rate on the validation set.