Search Results (359)

The economic impact of food fraud is difficult to quantify precisely, because fraud is structurally designed to evade detection; available estimates are indirect projections rather than direct forensic accounting and are commonly cited in the range of USD 10–15 billion annually. The integration of artificial intelligence (AI) with analytical instrumentation has generated a rapidly expanding body of research aimed at detecting adulteration, mislabeling, and substitution across food matrices. This systematic review examines the extent to which AI-assisted instrumental technologies contribute to food fraud prevention (as distinct from laboratory detection) and characterizes the structural factors that constrain real-world translation. A systematic search of the peer-reviewed literature published between 2021 and 2026 yielded 83 eligible records (80 primary studies and 3 review articles) after applying predefined inclusion criteria. Data were extracted into a structured seven-sheet workbook covering study characteristics, instrumental technologies, AI architectures, performance metrics, industrial-validation status, implementation evidence, and methodological quality. The corpus shows consistently high reported analytical accuracy under controlled laboratory conditions (median of extractable classification accuracies ≈ 99–100%; ≥95% in 86% of studies with an extractable value). At the same time, 68 of 83 studies (82%) reported no external validation, no study (0/83) achieved inter-laboratory validation, no study documented routine-monitoring application, and only one study reported testing in a genuine industrial environment. The most frequently featured platforms were NIR spectroscopy and electronic-nose arrays (each featuring in 30/83 studies, frequently in data-fusion combinations), followed by gas-chromatography-based systems (16/83) and hyperspectral imaging (13/83). Classical machine learning predominated (57/83 studies coded as classical ML, with a further 11 hybrid ML/DL designs and 12 deep-learning-only designs). A direct statistical comparison found no significant difference in reported accuracy between classical-ML and deep-learning studies (median 100% vs. 98.2%; Mann–Whitney U test, p = 0.16). A pre-specified test of the hypothesis that high reported accuracy is itself a marker of overfitting was not supported by the corpus: reported accuracy was not negatively associated with external-validation status (Fisher’s exact p = 0.51) or with methodological-quality score (Spearman ρ = 0.15, p = 0.23). Methodological quality was predominantly moderate (49/83 scored 3/5; 22 scored 2/5; 11 scored 4/5; one study scored 5/5), and 19/83 (23%) carried a high risk of bias. The review’s central observation—a measurable gap between demonstrated laboratory detection and evidenced real-world prevention—is well supported by the deployment, inter-laboratory, and routine-monitoring data. We deliberately separate this strongly evidenced conclusion from weaker inferences (e.g., the overfitting hypothesis) that the corpus cannot currently establish, and we outline a validation-driven, deployment-oriented research agenda. Full article

Economically motivated adulteration of powdered ginger with low-cost cereal flours represents an increasing concern for food authenticity and quality control. The aim of this study was to develop and validate a rapid, reliable, and non-destructive method for the quantitative determination of powdered ginger adulteration using visible and near-infrared (Vis–NIR) spectroscopy coupled with chemometric modelling. Ginger powder samples were adulterated with wheat, corn, and rice flours at concentrations ranging from 5 to 50% (w/w), with particular emphasis on the low-to-medium adulteration interval (10–25%), where reliable quantification is especially relevant for food fraud detection. Spectral data acquired in the visible (400–700 nm), near-infrared (700–2500 nm), and combined Vis–NIR (400–2500 nm) regions were preprocessed using Savitzky–Golay filtering and evaluated using Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), and Random Forest Regression (RFR). In addition, Linear Discriminant Analysis (LDA), Support Vector Machines (SVM), and Random Forest (RF) were compared for sample classification. Among the evaluated approaches, LDA achieved the highest classification accuracy (>95%) using the NIR spectroscopic region, while PLSR models developed from the NIR spectral region provided the best quantitative performance, with validation coefficients of determination above 0.99, prediction errors below 1%, and RPD values greater than 13. The results demonstrate that Vis–NIR spectroscopy combined with chemometric modelling enables accurate discrimination between authentic and adulterated samples, as well as reliable quantification of flour adulteration in powdered ginger without sample preparation or chemical reagents. The proposed methodology constitutes a rapid, environmentally friendly, and cost-effective analytical strategy with strong potential for routine quality control and food fraud prevention. Full article

Adulteration and contamination represent an ever-growing global problem, due to the emergence of increasingly sophisticated illicit commercial practices involving high-value and widely consumed products. Traditional chromatographic methods, such as liquid chromatography coupled with mass spectrometry, offer high sensitivity but their application is often limited by long analysis time, complex sample preparation, and the use of non-eco-friendly organic solvents. In contrast, enzymatic and immunoassay-based methods are rapid and require minimal sample preparation, but their applicability may be restricted by antibody specificity and potential cross-reactivity. In this context, near-infrared (NIR) spectroscopy is a rapid, non-destructive technique that does not require the use of solvents and is becoming increasingly relevant in food analysis. This review provides an overview of the potential of this spectroscopic technique, coupled with chemometrics, for the detection of adulterants and contaminants in various matrices. After a brief summary of the principles on which it is based and the instruments that can be used, the chemometric processes useful for data interpretation have been discussed, as well as the main applications in liquid and solid food, including oils, milk, juices, spices, cereals, coffee, and dietary supplements. The analyzed case studies indicated that even very low levels (ppm) of adulterants and contaminants can be detected with high accuracy and sensitivity using models such as PLSR, SVM, RF, and CNNs and new and emerging devices such as portable ones. Full article

Background/Objectives: Food irradiation is a recognized method for the microbiological decontamination and preservation of plant raw materials; however, in the European Union, its use is subject to strict legal regulation and to a mandatory labelling obligation. For dietary supplements, which are legally classified as foodstuffs, this issue is particularly important for market transparency and the consumer’s right to information. The aim of this pilot study was to assess the presence of post-irradiation characteristics in selected dietary supplements available at retail in Poland, with particular attention to products containing valerian root (Valeriana officinalis L.), and to compare the results obtained with the information declared on the product labels. Methods: The analysis included 15 analytical samples representing 11 commercial formulations. The study was performed using the thermoluminescence method according to standard PN-EN 1788:2002 in an ISO/IEC 17025-accredited laboratory. Results: Thermoluminescence characteristics consistent with prior irradiation were observed in 7 samples, whereas 8 samples did not show such properties. Positive results concerned exclusively products containing valerian root, including replicate batches of two commercial formulations. None of the positive samples carried any label information about irradiation or exposure to ionizing radiation. The results indicate that some dietary supplements on the market may be subjected to irradiation without this being properly reflected in their labelling. Conclusions: Given the pilot nature of the study and the limited, non-random sample, the results should be treated as a premise for further analysis rather than as a basis for generalizing to the entire market. The results of this study indicate a need for broader analytical monitoring and targeted control of dietary supplements containing plant raw materials. Full article

This preliminary study demonstrates the use of ultraviolet–visible (UV–vis) absorption spectroscopy coupled with chemometric modeling to detect and quantify honey adulteration. UV–vis spectra of pure honey and samples experimentally adulterated with corn, agave, and date syrups were collected within the 200–500 nm range after their simple dilution with water. Orthogonal partial least squares discriminant analysis (OPLS-DA) distinguished pure honey from adulterated samples, achieving 88% accuracy on an external validation dataset. A second set of classification models differentiated between honey containing a single adulterant and that containing multiple adulterants, with a high prediction rate based on an external validation dataset. Quantitative analysis of adulterant content via orthogonal partial least squares regression analysis revealed the strong predictive performance of the models (R² ≥ 0.87) on an external validation dataset. The proposed proof-of-concept study provides a rapid, cost-effective, and non-destructive screening tool for honey authentication with minimal sample preparation. Full article

Economically motivated adulteration of olive oil, coffee and fruit juice is a persistent food-fraud problem for which Fourier-transform infrared (FTIR) spectroscopy with chemometrics offers rapid screening. Linear partial least squares (PLS) is interpretable but cannot capture non-linear mixing; neural networks add flexibility at the cost of becoming black boxes. We evaluated a Kolmogorov–Arnold network (KAN), which places learnable univariate functions on its edges and is therefore intrinsically interpretable, against PLS, support-vector regression, random forests, a multilayer perceptron and a one-dimensional convolutional network on three attenuated total reflectance (ATR)–FTIR datasets (olive oil + sunflower oil, coffee + malt flour, orange juice + apple juice; approximately 350, 400 and 400 spectra). All models were compared under identical, leakage-free validation that splits spectra by physical sample. The compact KAN was consistently competitive (cross-validated coefficients of determination (R²) = 0.86, 0.93 and 0.69) and yielded closed-form equations whose variables map to recognised vibrational bands and whose importance ranking agrees with SHapley Additive exPlanations (SHAP; Spearman ρ = 0.86–0.90); symbolic conversion costs no accuracy. We also report the following limits: PLS was strongest where the chemistry was linear (coffee) and the multilayer perceptron was strongest on fruit juice, whose equation is the weakest (R² = 0.47–0.75 across seeds); a parameter-matched perceptron matched the KAN’s accuracy; and leave-one-brand-out validation degraded every model. The KAN is therefore a promising, compact and genuinely transparent alternative under controlled multi-matrix conditions, not a deployment-ready method. Full article

Fraud in herbal medicines and botanical supplements—such as biological substitution, dilution with inert material, undeclared active pharmaceutical ingredients (APIs), and intentional mislabeling—represents a significant and underestimated threat to public health. Beyond regulatory problems, fraudulent botanical products also compromise the scientific integrity of bioactive compound research. If pharmacological or clinical studies are carried out on adulterated material without knowing it, the resulting data on safety and efficacy become unreliable, and any conclusions drawn from such data are questionable. This review critically evaluates multi-platform metabolomics as an analytical decision-making framework for botanical fraud detection. Rather than cataloging available methods, we focus on the complete analytical pipeline: study design and sample strategy, multi-platform fingerprint acquisition and processing (NMR, GC-MS, LC-HRMS, HPTLC), and chemometric modeling, validation, and decision support. Particular emphasis is placed on data fusion across platforms and the requirements for producing decision-grade evidence—analytical outputs robust enough to support market monitoring and regulatory action. Full article

Hidden adulteration of stale berries in dried goji berry batches is difficult to detect by manual inspection or batch-level quality assessment. This study developed a high-throughput method for kernel-level spoilage adulteration quantification in dried goji berries. It addressed three practical challenges in the image processing of densely arranged dried-fruits, including scalable label generation for deep learning segmentation without pixel-level manual annotation, separation of densely touching small berries, and full-size quality level distribution map reconstruction. SAM-assisted pseudo-label generation combined with multi-scale image cropping was used to overcome the limitation of manual pixel-level annotation, while YOLO-based instance segmentation was further employed for efficient berry localization in dense scenes. The freshness labels of segmented single berries were assigned by a statistical RGB-HSV grading rule. Specifically, adaptive multi-scale image cropping for segmentation was applied to improve local separability of berries under dense adhesion and occlusion conditions. The crop-level segmentation and grading outputs were subsequently reconstructed into the original image coordinate system to generate complete quality distribution maps. Results showed that YOLO models trained based on the pseudo-labels achieved a precision of 0.953, a recall of 0.951, an mAP₅₀ of 0.960, and an mAP_50-95 of 0.846. The full-size grading map reconstruction method produced a mean duplicate-suppression rate of 4.31%. In the full freshness-grading test dataset, 4850 berries were detected, including 449 stale berries. The mean absolute counting error was 1.61%. The proposed framework reduces manual annotation requirements while enabling berry-level freshness classification and quantitative stale-berry proportion estimation, providing objective information for dried fruit quality screening and adulteration control. Full article

Background: Global food supply chains have become increasingly complex, sourcing ingredients from multiple countries and intermediaries, creating opportunities for fraud, adulteration, and mislabeling that may compromise consumer safety and market confidence. Digital traceability technologies have been suggested as potential countermeasures, but there is little concrete evidence of their impact in practice. This research presents an assessment of the maturity and effectiveness of these technologies, identifies implementation barriers and security/privacy concerns, and maps research gaps/future directions. Methods: A scoping review of 64 studies from 2023 to 2026 with data extracted from the Scopus and IEEE databases was carried out according to the PRISMA-ScR guidelines and a structured pre-specified data extraction framework. Results: The field is empirically immature, with none of the reviewed solutions offering a provably correct, adversarially tested solution to the oracle problem. There is a lack of alignment between on-chain immutability and GDPR right to erasure and an unequal burden of implementation costs imposed on smallholder producers. Conclusions: A gradual implementation of traceability regulations, along with cost-of-ownership models and harmonized certification measures that do not disadvantage smaller producers are proposed. Finally, field trials, adversarial testing and reporting results in a standardized format, capturing detection performance and implementation costs, are essential. Full article

Food products containing plant ingredients are particularly vulnerable to economically motivated adulteration (EMA), which poses risks to consumer trust and regulatory compliance. While traditional methods—such as microscopy, chemical profiling or targeted PCR—struggle to detect adulterants in processed food products or complex mixes, DNA metabarcoding offers a non-targeted, high-throughput alternative. This study presents a nanopore sequencing-based technique that is easy to implement, cost-effective and sufficiently sensitive to detect substitutions, with a focus on spices and herbal teas as model matrices. The method was evaluated using eight single-species reference samples and five commercial multi-ingredient products. It reliably detected undeclared contaminants (e.g., mint in oregano) and species substitutions. Compared to single-barcode approaches, the combination of ITS2 + matK + trnH-psbA markers achieved higher sensitivity. The proposed workflow requires minimal infrastructure and a 2–4-day turnaround time. However, factors such as DNA degradation in highly processed foods, database gaps, and biological diversity limited detection in some cases. These findings demonstrate the workflow’s potential as a first-line screening tool for food authenticity testing, aligning with requirements such as EU regulation 1169/2011 on food labelling or the FDA’s Economically Motivated Adulteration (EMA) program. Future work should validate the method against regulatory thresholds and expand testing to a broader variety of species and matrices. Full article

Synthetic azo dyes may be fraudulently added to paprika powder to intensify color, creating a need for rapid, at-line screening before confirmatory chromatographic analysis. This study evaluated handheld laser-induced breakdown spectroscopy (LIBS) for the screening of sodium-salt azo dye adulteration represented by Allura Red (E129), Ponceau 4R (E124), and Orange II in paprika powder. Paprika was spiked at 17 levels (0–6% w/w) in three independent batches, mixed with Al₂O₃ binder, dried, pelletized, and measured with a portable SciAps Z-903 under argon in the broad 190–950 nm region. The calibration models were developed using SNV-pretreated spectra and partial least squares regression with test-set validation based on a 70/30 split by concentration level. The main concentration-related response was increased Na emission, consistent with the sodium-salt form of the dyes, while Al emission showed an inverse matrix-/plasma-coupled trend despite the constant binder fraction. Full-spectrum models gave R²_TSV values of 0.845–0.875 and RMSETSV values of 0.638–0.716% (w/w), using 4–5 latent variables. Outlier trimming improved prediction for Allura Red and Orange II, while regression coefficient-uncertainty variable selection reduced model complexity to 2–3 latent variables with dye-dependent effects. Since sodium-salt azo dyes are commonly used as adulterants in paprika powder, the Na-dominated LIBS response demonstrates good potential for rapid at-line screening. Nevertheless, as LIBS does not directly confirm dye identity, complementary confirmatory analysis using spectroscopic or chromatographic methods remains necessary. Full article

Honey is a high-value food product that is vulnerable to adulteration with exogenous sugars, posing challenges for food authenticity and consumer protection. This study applied Stable Isotope Ratio Analysis (SIRA) to assess the authenticity of honey collected from three major honey-producing regions of Ghana (Volta, Bono and Bono East). A total of 28 honey samples were analysed by elemental analysis–isotope ratio mass spectrometry (EA-IRMS) to obtain carbon (δ¹³C), nitrogen (δ¹⁵N) and sulphur (δ³⁴S) isotope composition. Honey authenticity was evaluated according to AOAC Official Method 998.12 by comparing δ¹³C values of bulk honey and the corresponding protein fraction. The δ¹⁵N and δ³⁴S values in honey protein were used to investigate environmental and regional variability. Samples without detectable C4 adulteration exhibited δ¹³C_protein values consistent with C3 floral sources, whereas several samples showed Δδ¹³C values more negative than −1.0‰, indicating the presence of C4-derived sugars above the AOAC adulteration threshold. Calculated C4 sugar contents ranged from 8 to 12% in moderately adulterated samples to as high as 78–79% in severely adulterated samples, confirming substantial dilution with C4 sugars. Nitrogen and sulphur isotope ratios provide additional information on environmental and regional variability among the sampled regions. Principal Component Analysis revealed that the first two principal components (PC1 and PC2) accounted for 83.8% of the total variance 83.8% of the total variance and showed separation between samples with detectable C4 adulteration and those without, while highlighting regional isotopic differences. These results demonstrate that stable isotope analysis is an effective tool for detecting C4 sugar adulteration in honey and that the combined use of carbon, nitrogen and sulphur isotopes can provide additional information on environmental and regional variability. These findings provide preliminary isotopic data on honey collected from three major honey-producing regions of Ghana and support the application of the stable isotope approach for honey authenticity assessment and quality control. Full article

Processed meat products represent a major challenge for proteomic species identification due to extensive thermal treatment and protein structural changes. In this study, species-specific peptides in pork, chicken, and bovine meat products were identified using a directed fragmentation-assisted de novo sequencing workflow that combines 4-formylbenzene-1,3-disulfonic acid (FBDA) peptide derivatization, dual-polarity data-independent mass spectrometry (DIA-MS), and Protein Acrobat de novo sequencing software. Comparative analysis of non-fractionated and strong cation exchange (SCX)-fractionated pork luncheon samples improved peptide and protein identification after fractionation, with 312 peptides and 115 protein groups detected exclusively in fractionated samples. Species-specific peptides were predominantly assigned to conserved muscle-related proteins, including myosin, troponin, and tropomyosin, while sequence variability enabled reliable species discrimination despite protein conservation across species. To evaluate applicability for food fraud detection, mixed meat samples containing 10% chicken in pork or bovine matrices were analyzed, reflecting potential economically motivated adulteration through substitution with lower-cost meat components. Several chicken-specific peptides remained detectable in both mixtures, demonstrating robustness of the FBDA-assisted peptide sequencing combined with SCX fractionation and DIA-MS for detection of adulteration in complex processed food matrices. These findings establish a mass spectrometry-driven orthogonal method to ELISA testing for fast, reliable and accurate metaproteome analysis of highly processed food. Full article

The integrity of halal supply chains is increasingly threatened by fragmented paper-based records, certificate fraud, and the absence of real-time traceability. This paper presents HalalChain, a blockchain-based halal product traceability system that enforces role-based access control (RBAC) through three Solidity smart contracts deployed on an Ethereum-compatible blockchain. HalalChain is designed for production deployment on an EVM-compatible Layer-2 or sidechain such as Polygon or BNB Chain, on which the contracts run without code changes. A dual-storage architecture synchronises every supply chain event to both a PostgreSQL relational database and the blockchain, balancing on-chain immutability with off-chain query performance. The system supports five stakeholder roles, namely administrator, supplier, manufacturer, logistics, and retailer, each restricted to specific supply chain event types enforced at the smart contract level. Consumers can verify product halal status and full supply chain history by scanning a QR code linked to a public verification endpoint that cross-checks database records against on-chain event counts, producing a chain-integrity indicator. As the current chain-integrity check is count-base, it can detect missing or extra database rows, but it cannot detect content-level modification if the row count remains unchanged. A total of 107 automated test cases were executed covering functional correctness, edge cases, end-to-end integration, and gas performance benchmarks. Core smart contract operations consume between 25,365 and 213,684 gas units, indicating feasible deployability on Ethereum-compatible networks. An exploratory analysis was carried out with a preliminary survey of 40 respondents (mean = 4.10 on a 5-point Likert scale), suggesting that consumer demand for blockchain-verified halal certification is encouraging. The results demonstrate that HalalChain provides a tamper-evident, role-enforced traceability foundation for the halal food industry. The system secures the digital chain of custody cryptographically and the physical–digital binding between the QR code, and the product remains a separate trust assumption requiring complementary anti-tamper mechanisms. Full article

This study presents a novel analytical approach combining headspace solid-phase microextraction (HS-SPME) with gas chromatography–mass spectrometry (GC–MS) and advanced chemometric techniques to detect adulteration in coriander honey. A total of 34 volatile compounds were identified and quantified, revealing a progressive decrease in both profile complexity and compound abundance with increasing levels of invert sugar and high-fructose corn syrup (HFCS) adulteration. Chromatographic and chemometric analyses effectively distinguished authentic from adulterated samples, with the Extreme Gradient Boosting (XGBoost) model achieving a high classification performance of 95.83% accuracy. The study highlights the critical impact of adulteration on honey’s chemical composition and confirms the efficacy of integrating modern analytical and machine learning tools for rapid, sensitive, and reliable honey authenticity assessment. This methodology offers a valuable framework for food quality control and fraud prevention, addressing current challenges in the honey market and protecting consumer interests. Full article