Rapid fraud detection of cocoa powder with carob flour using near infrared spectroscopy

Statistical machine learning techniques applied to NIR spectral data for rapid detection of sudan dye-I in turmeric powders with optimized pre-processing and wavelength selection

Machine learning techniques were applied systematically to the spectral data of near-infrared (NIR) spectroscopy to find out the sudan dye I adulterants in turmeric powders. Turmeric powder is one of the most commonly used spice and a simple target for adulteration. Pure turmeric powder was prepared at the laboratory and spiked with sudan dye I adulterants. The spectral data of these adulterated mixtures were obtained by NIR spectrometer and investigated accordingly. The concentrations of the adulterants were 1%, 5%, 10%, 15%, 20%, 25%, 30% (w/w) respectively. Exploratory data analysis was done for the visualization of the adulterant classes by principal component analysis (PCA). Optimization of the pre-processing and wavelength selection was done by cross-validation techniques using a partial least squares regression (PLSR) model. For quantitative analysis four different regression techniques were applied namely ensemble tree regression (ENTR), support vector regression (SVR), principal component regression (PCR), and PLSR, and a comparative analysis was done. The best method was found to be PLSR. The accuracy of the PLSR analysis was determined with the coefficients of determination (R2) of greater than 0.97 and with root mean square error (RMSE) of less than 0.93 respectively. For the verification of the robustness of the model, the Figure of merit (FOM) of the model was derived with the help of the Net analyte signal (NAS) theory. The current study established that the NIR spectroscopy can be applied to detect and quantify the amount of sudan dye I adulterants added to the turmeric powders with satisfactory accuracy.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13197-024-05971-9.

Detection and quantification of cocoa powder adulteration using Vis-NIR spectroscopy with chemometrics approach

The rising demand for cocoa powder has resulted in an upsurge in market prices, leading to the emergence of adulteration practices aimed at achieving economic benefits. This study aimed to detect and quantify cocoa powder adulteration using visible and near-infrared spectroscopy (Vis-NIRS). The adulterants used in this study were powdered carob, cocoa shell, foxtail millet, soybean, and whole wheat. The NIRS data could not be resolved using Savitzky-Golay smoothing. Nevertheless, the application of a random forest and support vector machine successfully classified the samples with 100% accuracy. Quantification of adulteration using partial least squares (PLS), Lasso, Ridge, elastic Net, and RF regressions provided R2 higher than 0.96 and root mean square error <2.6. Coupling PLS with the Boruta algorithm produced the most reliable regression model (R2 = 1, RMSE = 0.0000). Finally, an online application was prepared to facilitate the determination of adulterants in the cocoa powder.

Authentication of Cocoa Products Based on Profiling and Fingerprinting Approaches: Assessment of Geographical, Varietal, Agricultural and Processing Features

Cocoa and its derivative products, especially chocolate, are highly appreciated by consumers for their exceptional organoleptic qualities, thus being often considered delicacies. They are also regarded as superfoods due to their nutritional and health properties. Cocoa is susceptible to adulteration to obtain illicit economic benefits, so strategies capable of authenticating its attributes are needed. Features such as cocoa variety, origin, fair trade, and organic production are increasingly important in our society, so they need to be guaranteed. Most of the methods dealing with food authentication rely on profiling and fingerprinting approaches. The compositional profiles of natural components -such as polyphenols, biogenic amines, amino acids, volatile organic compounds, and fatty acids- are the source of information to address these issues. As for fingerprinting, analytical techniques, such as chromatography, infrared, Raman, and mass spectrometry, generate rich fingerprints containing dozens of features to be used for discrimination purposes. In the two cases, the data generated are complex, so chemometric methods are usually applied to extract the underlying information. In this review, we present the state of the art of cocoa and chocolate authentication, highlighting the pros and cons of the different approaches. Besides, the relevance of the proposed methods in quality control and the novel trends for sample analysis are also discussed.

Rapid determination of the shell content in cocoa products using FT-NIR spectroscopy and chemometrics

The determination of the cocoa shell content is of interest because a high shell content causes a reduction in the quality of cocoa products. Consequently, the aim of the present study was the development of a routinely applicable method for the quantitation of shell material in cocoa nibs. For this, 51 fermented cocoa samples of different varieties from 14 cocoa growing countries covering the crop years 2012-2017 were acquired. Admixtures of cocoa nibs with shell material were prepared in a range of 0-20% cocoa shell and subsequently analysed by Fourier transform near-infrared spectroscopy (FT-NIRS). Support vector machine regression models were created, which enabled the prediction of the cocoa shell content in a mixing ratio range of 0-20% with an RMSE of 2.05% and a R² of 0.88 and in a range of 0-10% with an RMSE of 1.70% and a R² of 0.72. This predictive capability suggests that the presented method is suitable for rapid determination of cocoa shell content in cocoa nibs. In addition, it was demonstrated that the method is applicable to other relevant cocoa matrices, as the prediction of the shell content of several industrial cocoa masses by the FT-NIRS-based model showed good consistency with the prediction by liquid chromatography-mass spectrometry. This emphasizes that FT-NIRS combined with chemometrics has great potential for the determination of cocoa shell content in cocoa nibs and cocoa masses in routine analysis, such as incoming inspection.

Handheld and benchtop vis/NIR spectrometer combined with PLS regression for fast prediction of cocoa shell in cocoa powder

The fermented and dried cocoa beans are peeled, either before or after the roasting process, as peeled nibs are used for chocolate production, and shell content in cocoa powders may result from economically motivated adulteration (EMA), cross-contamination or misfits in equipment in the peeling process. The performance of this process is carefully evaluated, as values above 5% (w/w) of cocoa shell can directly affect the sensory quality of cocoa products. In this study chemometric methods were applied to near-infrared (NIR) spectra from a handheld (900-1700 nm) and a benchtop (400-1700 nm) spectrometers to predict cocoa shell content in cocoa powders. A total of 132 binary mixtures of cocoa powders with cocoa shell were prepared at several proportions (0 to 10% w/w). Partial least squares regression (PLSR) was used to develop the calibration models and different spectral preprocessing were investigated to improve the predictive performance of the models. The ensemble Monte Carlo variable selection (EMCVS) method was used to select the most informative spectral variables. Based on the results obtained with both benchtop (R²_P = 0.939, RMSE_P = 0.687% and RPD_P = 4.14) and handheld (R²_P = 0.876, RMSE_P = 1.04% and RPD_P = 2.82) spectrometers, NIR spectroscopy combined with the EMCVS method proved to be a highly accurate and reliable tool for predicting cocoa shell in cocoa powder. Even with a lower predictive performance than the benchtop spectrometer, the handheld spectrometer has potential to specify whether the amount of cocoa shell present in cocoa powders is in accordance with the Codex Alimentarius specifications.

Automatic and non-targeted analysis of the volatile profile of natural and alkalized cocoa powders using SBSE-GC-MS and chemometrics

A total of 56 key volatile compounds present in natural and alkalized cocoa powders have been rapidly evaluated using a non-target approach using stir bar sorptive extraction gas chromatography mass spectrometry (SBSE-GC-MS) coupled to Parallel Factor Analysis 2 (PARAFAC2) automated in PARADISe. Principal component analysis (PCA) explained 80% of the variability of the concentration, in four PCs, which revealed specific groups of volatile characteristics. Partial least squares discriminant analysis (PLS-DA) helped to identify volatile compounds that were correlated to the different degrees of alkalization. Dynamics between compounds such as the acetophenone increasing and toluene and furfural decreasing in medium and strongly alkalized cocoas allowed its differentiation from natural cocoa samples. Thus, the proposed comprehensive analysis is a useful tool for understanding volatiles, e.g., for the quality control of cocoa powders with significant time and costs savings.

Amazonian cacao-clone nibs discrimination using NIR spectroscopy coupled to naïve Bayes classifier and a new waveband selection approach

Near-Infrared Spectroscopy (NIRS) has shown to be helpful in the study of rice, tea, cocoa, and other foods due to its versatility and reduced sample treatment. However, the high complexity of the data produced by NIR sensors makes necessary pre-treatments such as feature selection techniques that produce compact profiles. Supervised and unsupervised techniques have been tested, creating different subsets of features for classification, which affect the performance of the classifiers based on such compact profiles. In this sense, we propose and test a new covering array feature selection (CAFS) algorithm coupled to the naïve Bayes classifier (NBC) to discriminate among Amazonian cacao nibs from six cacao clones. The CAFS wrapper approach looks for the wavebands that maximize the F₁-score, and then, are more relevant for classification. For this purpose, cacao pods of six varieties were collected, and their grains were extracted and processed (fermented, dried, roasted, and milled) to obtain cacao nibs. Then from each clone NIR spectral profiles in the range of 1100-2500 nm were extracted, and relevant wavebands were selected using the proposed CAFS algorithm. For comparison, two standard feature selection techniques were implemented the multi-cluster feature selection MCFS and the eigenvector centrality feature selection ECFS. Then, based on the different selected variables, three NBCs were built and compared among them through statistical metrics. The results showed that using the wavebands selected by CAFS, the NBC performed an average accuracy of 99.63%; being this superior to the 94.92% and 95.79% for ECFS and MCFS respectively. These results showed that the wavebands selected by the proposed CAFS algorithm allowed obtaining a better fit concerning other feature selection methods reported in the literature.

Determination of adulteration in wheat flour using multi-grained cascade forest-related models coupled with the fusion information of hyperspectral imaging

Wheat flour (WF) is a common ingredient in staple foods. However, the presence of intentional or unintentional adulterants makes it difficult to guarantee WF quality. Multi-grained cascade forest (gcForest) model, a non-neural network deep learning structure, fused with image-spectra features from hyperspectral imaging (HSI) was employed for detecting adulterant type (peanut, walnut, or benzoyl peroxide) and the corresponding concentration (0.03%, 0.05%, 0.1%, 0.5%, 1%, and 2%). Based on the spectra of full wavelength and effective wavelength (EW) from hyperspectral images of WF samples, the gcForest-related models exhibited high performance (lowest ACC_P = 92.45%) and stability (lowest area under the curve = 0.9986). Furthermore, the fusion of the EW and the image features extracted by the symmetric all convolutional neural network (SACNN) was used to establish the gcForest-related models. The maximum accuracy improvement of the fusion feature model relative to the single spectral model and the image model was 2.45% and 44.37%, respectively. The results indicate that the gcForest-related model, combined with the image-spectra fusion feature of HSI, provides an effective tool for detection in food and agriculture.

Differentiation of Organic Cocoa Beans and Conventional Ones by Using Handheld NIR Spectroscopy and Multivariate Classification Techniques

The global market for organic cocoa beans continues to show sturdy growth. A low-cost handheld NIR spectrometer (900-1700 nm) combined with multivariate classification algorithms was used for rapid differentiation analysis of organic cocoa beans' integrity. In this research, organic and conventionally cultivated cocoa beans were collected from different locations in Ghana and scanned nondestructively with a handheld spectrometer. Different preprocessing treatments were employed. Principal component analysis (PCA) and classification analysis, RF (random forest), KNN (K-nearest neighbours), LDA (linear discriminant analysis), and PLS-DA (partial least squares-discriminant analysis) were performed comparatively to build classification models. The performance of the models was evaluated by accuracy, specificity, sensitivity, and efficiency. Second derivative preprocessing together with PLS-DA algorithm was superior to the rest of the algorithms with a classification accuracy of 100.00% in both the calibration set and prediction set. Second derivative algorithm was found to be the best preprocessing tool. The identification rates for the calibration set and prediction set were 96.15% and 98.08%, respectively, for RF, 91.35% and 92.31% for KNN, and 90.38% and 98.08% for LDA. Generally, the results showed that a handheld NIR spectrometer coupled with an appropriate multivariate algorithm could be used in situ for the differentiation of organic cocoa beans from conventional ones to ensure food integrity along the cocoa bean value chain.

Front-face synchronous fluorescence spectroscopy for rapid and non-destructive determination of free capsanthin, the predominant carotenoid in chili (Capsicum annuum L.) powders based on aggregation-induced emission

Capsanthin is the major natural carotenoid pigment in red chili pepper possessing important bioactivity. Its conventional determination method is high performance liquid chromatography (HPLC) with complex and tedious sample pretreatment. In this study, synchronous front-face fluorescence spectroscopy (FFFS) was applied for the fast and non-invasive detection of free capsanthin in chili powders. Although capsanthin was only weak fluorescent in solution state, it showed strong fluorescence in two separated regions in front-face geometry which could also be clearly observed in chili powders. The mechanisms of these emissions are revealed to be aggregation-induced emission (AIE) and J-aggregate formation (JAF). The free capsanthin in 85 chili powder samples were determined by HPLC as in the range of 0.6-3.0 mg/g. The total synchronous FFFS spectra of these samples were scanned. Simple first-order models were built by partial least square regression (PLSR), and were validated by 5-fold cross-validation and external validation. The coefficients of determination (R²) were higher than 0.9, and the root mean square errors (RMSE) were less than 0.2 mg/g. The relative error of prediction (REP) was 9.9%, and the residual predictive deviation (RPD) was 3.7. The method was applied for the estimation of free capsanthin in several real-world samples with satisfactory analytical results. The average relative error to HPLC reference values was -11.8%.

NIR and MIR spectroscopy for quick detection of the adulteration of cocoa content in chocolates

The Near (NIR) and Mid (MIR) Infrared Spectroscopy associated with chemometric techniques were used to determine the cocoa solids content in chocolates and detect possible adulterations. Five chocolate formulations (30% to 90%) were produced with different cocoa solids concentrations and 110 commercial samples from 10 different countries with varying concentrations of cocoa solids (30% to 88%) were acquired. All repetions of the produced and commercial chocolates were evaluated using NIR and MIR. Spectroscopic data were submitted to multivariate techniques of Principal Component Analysis (PCA) and Partial Least Squares Regression (PLS). For both spectroscopy techniques, the PCA of the 5 formulations formed 5 distinct groups regarding the cocoa solids and the commercial samples showed a behavior pattern similar to the produced samples. For PLS, the regression equations showed high predictive capacity, with correlation coefficients above 90 and RMSECV values of 0.70 and 1.22, for NIR and MIR, respectively. These models highlighted, approximately, 14% of the commercial samples as possible adulterated products.

Rapid Authentication of 100% Italian Durum Wheat Pasta by FT-NIR Spectroscopy Combined with Chemometric Tools

Italy is the country with the largest durum wheat pasta production and consumption. The mandatory labelling for pasta indicating the country of origin of wheat has made consumers more aware about the consumed pasta products and is influencing their choice towards 100% Italian wheat pasta. This aspect highlights the need to promote the use of domestic wheat as well as to develop rapid methodologies for the authentication of pasta. A rapid, inexpensive, and easy-to-use method based on infrared spectroscopy was developed and validated for authenticating pasta made with 100% Italian durum wheat. The study was conducted on pasta marketed in Italy and made with durum wheat cultivated in Italy (n = 176 samples) and on pasta made with mixtures of wheat cultivated in Italy and/or abroad (n = 185 samples). Pasta samples were analyzed by Fourier transform-near infrared (FT-NIR) spectroscopy coupled with supervised classification models. The good performance results of the validation set (sensitivity of 95%, specificity and accuracy of 94%) obtained using principal component-linear discriminant analysis (PC-LDA) clearly demonstrated the high prediction capability of this method and its suitability for authenticating 100% Italian durum wheat pasta. This output is of great interest for both producers of Italian pasta pointing toward authentication purposes of their products and consumer associations aimed to preserve and promote the typicity of Italian products.

Discrimination of Tetrastigma hemsleyanum according to geographical origin by near-infrared spectroscopy combined with a deep learning approach

Recently, deep learning has presented as a powerful approach to overcome the deficiencies of the conventional biochemical approaches. In this study, a method for discriminating medicinal plant Tetrastigma hemsleyanum from different origins was proposed using near-infrared spectroscopy (NIRS) and deep learning models. Support vector machine (SVM), self-adaptive evolutionary extreme learning machine (SAE-ELM), and convolutional neural network (CNN) were used to process the near-infrared spectral data (4000-5600 cm^-1). The results indicated that the average recognition accuracy of SVM on the test set samples (n = 60) reached 90%. The average recognition accuracy of SAE-ELM was 98.3%, while CNN correctly discriminated 100% of T. hemsleyanum from different origins. Notably, CNN avoids tedious redundant data preprocessing and is also able to save the trained model for the next call to achieve rapid detection. As above, this study provides an effective deep learning-based method for discriminating the geographical origins of T. hemsleyanum as well as providing a convenient and satisfactory approach to ensure the famous-region of other medicinal plants.

Detection of fraud in high-quality rice by near-infrared spectroscopy

A key feature of food fraud is the use of a lower value ingredient to imitate an authentic product. This study was based on near-infrared spectroscopy (NIRS) analysis technology, partial least squares discriminant analysis (PLS-DA), and a support vector machine (SVM) to detect whether high-quality rice was mixed with other varieties of rice. As an aid to qualitative discrimination, PLS was used to establish the quantitative analysis model to assist in the recognition of the degree of fraud. Due to the direct correlation between the results of NIRS analysis and the homogeneity of the samples, four groups of samples with different physical forms (full granules, 40 mesh, 70 mesh, and 100 mesh) were prepared, each group consisted of 20 pure samples and 140 mixed samples, and the mixing ratio was between 5% and 50%, with an interval of 5%. Regarding qualitative analysis, the performance of the model has no obvious relationship with the physical state of the sample, the qualitative model of PLS-DA and SVM can detect the fraudulent rice with a 5% detection limit, respectively. Regarding quantitative analysis, the performance of the prediction model was closely related to the particle size of the samples: 100 mesh > 70 mesh > 40 mesh > full grains. The determination coefficient and root mean square errors of the optimal prediction result were 0.96 and 2.93, respectively. These results demonstrate that NIRS analysis technology is a reliable and fast tool to determine whether high-quality rice contains other varieties of rice. PRACTICAL APPLICATION: The work of this article is based on the current background of increasingly serious rice fraud, using near-infrared spectroscopy to quickly identify fraudulent rice, to a certain extent, and effectively alleviate the rice fraud. This technology can serve for the supervision of food regulatory agencies on rice fraud, and can also be used in food factories to ensure the authenticity of raw materials of rice.

Chocolate Quality Assessment Based on Chemical Fingerprinting Using Near Infra-red and Machine Learning Modeling

Chocolates are the most common confectionery and most popular dessert and snack across the globe. The quality of chocolate plays a major role in sensory evaluation. In this study, a rapid and non-destructive method was developed to predict the quality of chocolate based on physicochemical data, and sensory properties, using the five basic tastes. Data for physicochemical analysis (pH, Brix, viscosity, and color), and sensory properties (basic taste intensities) of chocolate were recorded. These data and results obtained from near-infrared spectroscopy were used to develop two machine learning models to predict the physicochemical parameters (Model 1) and sensory descriptors (Model 2) of chocolate. The results show that the models developed had high accuracy, with R = 0.99 for Model 1 and R = 0.93 for Model 2. The thus-developed models can be used as an alternative to consumer panels to determine the sensory properties of chocolate more accurately with lower cost using the chemical parameters.