Food Analytical Methods最新文献

This study examines the efficacy of natural deep eutectic solvents (NADES) in conjunction with ultrasound-assisted extraction (UAE) for extracting polyphenols from Vernonia amygdalina leaves. Optimization via central composite design (CCD) identified a NADES formulation (1:1.413 molar ratio of choline chloride to glucose with 79.125% water) that maximized total phenolic content (TPC) to 163.50 mg GAE/mL. This environmentally sustainable extraction technique considerably surpassed traditional ethanol-based extraction. The characterization using LC-HRMS, TEM, FTIR, and DSC validated a diverse phenolic profile, spherical particle shape, stable hydrogen-bonding interactions, and thermal stability of the optimized extract. The antioxidant evaluation revealed significant efficacy, achieving up to 89.27% DPPH inhibition at a 20% concentration. The anti-inflammatory properties evaluated in LPS-induced RAW 264.7 cells demonstrated a decrease in nitric oxide, reactive oxygen species, and proinflammatory cytokine production, highlighting the extract’s therapeutic potential. This work emphasizes the combined advantages of sustainable green extraction techniques and the potential biological activity of V. amygdalina, facilitating its use in natural therapies and nutraceuticals.

Honey consumption is increasing worldwide due to its natural qualities, but its safety and authenticity are not always assured. Honey adulteration through mislabeling and fraudulent production significantly damages consumer trust and poses health risks. This study aimed to identify the botanical makeup of Kazakh and imported Honey to assess whether it matches with the declared labeling using melissopalynological analysis and DNA metabarcoding. During our research, melissopalynological analysis showed that 64.7% of Kazakh Honey samples and 62.5% of imported Honey matched the declared labeling. In 6 samples of domestic Honey and 6 samples of imported honey, the plant source differed from what was indicated on the label. According to DNA metabarcoding data, the botanical origin of honey by the ITS2 marker was confirmed in 58.8% of Kazakh and 31.25% of imported honey samples. Using the rbcL marker, only 29.4% of domestic and 31.25% of imported honey samples matched the declared plant origin. Agreement between melissopalynology and ITS2 marker data was observed in 10 samples, and between melissopalynology and rbcL in 8 samples. Eight samples also matched the labeling based on both genetic markers. Thus, combining DNA metabarcoding with melissopalinological analysis enables a more accurate identification of the honey's botanical origin, thereby strengthening product authenticity control, helping to detect counterfeits, facilitating access to international markets, and increasing consumer confidence. Overall, the results highlight the importance of strict monitoring and verification methods to protect consumers and maintain market integrity.

Tryptophan, an essential amino acid for the human body, plays a crucial role in various physiological processes, including protein synthesis, fatty acid metabolism, enhancing energy levels, and being converted into neurotransmitters to regulate sleep and mood. It is also used in the prevention and treatment of pellagra. However, excessive intake of tryptophan can cause adverse effects, so detecting its concentration is of great significance. In this paper a silicon-based carbon quantum dot (N,Si-CDs) tryptophan probe was prepared. The structure and composition of N,Si-CDs were characterized by TEM, FT-IR, and XPS. Fluorescence studies revealed that N,Si-CDs exhibited high selectivity for tryptophan with minimal interference from other substances. There was a good Linear relationship between the fluorescence intensity of N,Si-CDs and the concentration of tryptophan in the range of 5–25 μM, and the detection Limit was as low as 19 nM. N,Si-CDs were successfully applied to detect tryptophan in complex Matrices such as honey, Yili milk powder, water-soluble vitamin C, sour plum soup, and lemon tea, with recovery rates ranging from 81.3% to 106.4%. Cell experiments demonstrated that N,Si-CDs had low toxicity to T24 cells and could be used for live-cell imaging. This N,Si-CDs shows great potential in fields such as food safety, environmental monitoring, and biomedical research.

The widespread use of monensin (MON) as an ionophore antibiotic in poultry production has raised significant concerns regarding residue accumulation in edible tissues, particularly given its potent cardiotoxic effects and narrow safety margins in Humans. Current analytical methods for MON detection primarily rely on expensive instrumentation, limiting accessibility for routine monitoring in resource-constrained regions with expanding poultry production. This study addresses the critical need for a cost-effective, field-deployable analytical platform by developing the first Dual-emission ratiometric fluorometric sensor specifically designed for MON residue detection in poultry tissues. The innovative sensing mechanism exploits the differential fluorescence responses of Rhodamine 6G (R6G) and red-emitting carbon dots (RCDs) upon MON interaction, where selective quenching of RCD emission (608 nm) occurs simultaneously with R6G signal enhancement (550 nm) under 525 nm excitation. This Dual-response system provides unprecedented analytical robustness through built-in internal calibration, eliminating matrix interference effects that plague conventional single-wavelength methods. The sensor demonstrated exceptional performance with linear detection ranges of 0.1–8.0 ng/mL in standard solutions and 5.0–400.0 ng/g in tissue matrices, achieving recovery efficiencies of 95.9–98.0% in muscle and 96.2–96.8% in liver samples. This platform uniquely enables real-time monitoring of MON elimination kinetics, facilitating evidence-based withdrawal period determination for specific production conditions. The methodology offers developing nations and small-scale producers an accessible compliance tool without requiring sophisticated instrumentation. Future applications include multiplex ionophore detection and portable device integration for on-site screening, potentially revolutionizing antibiotic monitoring in global food supply chains.

Graphical Abstract

This study proposes a suspension analysis approach for determining total carbohydrates in beans using spectrophotometry. Although previous studies have quantified carbohydrates using similar colorimetric reactions—typically after hydrolysis or extraction steps—this method employs a minimal sample preparation strategy, applied to intact bean matrices, eliminating the need for time-consuming procedures. Thirty-six bean samples were processed through grinding and sieving before suspension in water, followed by a colorimetric reaction with sulfuric acid and phenol. Validation results like linearity (correlation coefficient of 0.9995), limit of quantification (5.3%), accuracy (comparison with four reference materials), and precision (relative standard deviation up to 10%) were satisfactory. A comparison between the determined carbohydrate levels and those indicated on the bean packaging revealed consistency for most samples analyzed. The contribution of beans to the Recommended Dietary Allowance of carbohydrates (12–13%) showed the importance of grain in the intake of this macronutrient, along with other sources of carbohydrates, like rice, corn, and potatoes.

Graphical Abstract

A practical and explainable deep learning (DL) framework for the multi-classification of fruit quality conditions is developed in this study. The proposed system provides a non-intrusive and scalable automated quality inspection solution for fruits, addressing concerns related to food adulteration and spoilage. This approach achieves high accuracy and interpretability by combining DL models with optimization and visual explanations based on Grad-CAMs, and it has potential applications in agriculture, retail, and food safety. The dataset comprises 10,160 images of five common fruits (apple, banana, grape, mango, and orange), with two quality classes (formalin-mixed, fresh, rotten) for each fruit and three quality classes for the rotten classification (formalin-mixed, fresh, rotten). The data is separated into 70% for training, 15% for validation, and 15% for testing to achieve good learning performance and generalization. To overcome the problem of overfitting, the sixfold cross-validation method is employed to validate its stability. This study analyzes four DL models, including Vision Transformer (ViT), InceptionResNetV2, RepVGG, and CoAtNet. The latter is compared with the method presented herein, SA-CoViT-XAI, which combines CoAtNet with simulated annealing (SA) for hyperparameter tuning and Grad-CAM for interpretability. The developed model is intended to possess both reliable predictive performance and interpretability features. All models performed well in classifying multiple classes in the multi-class dataset. The proposed SA-CoViT-XAI model achieved the best accuracy, precision, recall, F1-score, and specificity of 99.61%, 99.71%, 99.60%, 99.65%, and 99.81%, respectively. It was also superior to the other models in terms of specificity, precision, recall, and F1 Score. The Grad-CAM visualizations demonstrated that the model successfully identifies and highlights regions of the fruit that are indicative of decay. The outcomes indicate that the SA-CoViT-XAI model is robust and explainable, and potentially applicable to real-world cases in fruit quality supervision, food detection, and grading. It uses simulated annealing for end-to-end training. Simultaneously, Grad-CAM offers rich information on decision-making; hence, the approach can be positioned as a valuable development in smart agriculture as well as AI-based food inspection.

Propolis is a bioactive resinous by-product produced by stingless bees to protect their colonies from environmental stressors and microbial threats. Rich in plant-derived phenolic compounds, its biological potential largely depends on effective extraction methods. This study evaluated the influence of ethanol concentration (20% and 80%) on the recovery of phenolic compounds and the biological properties of propolis from three stingless bee species: Tetragonisca angustula (Jataí), Melipona quadrifasciata (Mandaçaia), and Plebeia droryana (Mirim). Total phenolic content (TPC) and antioxidant activity (AA) were assessed using ABTS, DPPH, and FRAP assays. Phenolic profiles were determined by HPLC–DAD, and antimicrobial activity was evaluated against Salmonella Typhimurium, Escherichia coli, Listeria monocytogenes, and Bacillus subtilis through minimum inhibitory (MIC) and bactericidal concentrations (MBC). Results indicated that M. quadrifasciata propolis extracted with 80% ethanol exhibited the highest TPC (61.02 mg GAE g⁻¹; GAE: gallic acid equivalent) and antioxidant potential (26.54 mM TE g⁻¹; TE: Trolox). Key phenolic compounds, including catechin, epicatechin, p-coumaric acid, ferulic acid, cinnamic acid, chrysin, and galangin, were identified and quantified. Moreover, 80% ethanolic extracts showed superior antimicrobial efficacy across all tested pathogens with an MIC of 5.44 µg mL⁻¹. In conclusion, propolis from stingless bees is a promising source of phenolic compounds with strong antioxidant and antimicrobial activities, highlighting its potential for applications in food preservation and health-related products.

Food allergy is a sensitivity to a food or one of its components that triggers the immune system to react. It affects around 8% of children and 10% of adults. This study investigates the use of near-infrared spectroscopy (NIRS) combined with artificial intelligence (AI) approaches for rapid and non-destructive detection of lipid transfer proteins (LTP), a cause of severe food allergy. Unlike traditional chemical detection methods, which are time-consuming, costly, and often require complex sample preparation, NIRS offers a fast, non-invasive alternative suitable for routine food safety screening. NIRS spectra were acquired from a wide variety of foods, both with and without LTP, using a miniature spectrometer. Three deep learning architectures, convolutional neural networks (CNN), vision transformers (ViT), and TabTransformer, were employed to classify foods according to the presence of LTP, with hyperparameters optimized via Bayesian optimization. The findings indicated that ViT and CNN-based models had significant potential, achieving accuracies and F1 scores above 90%. Key wavelengths in the 1325–1455 nm range were identified as useful for identifying foods with LTP, reflecting changes in water, fat, and protein content. Additionally, AI explainability methods (SHAP, LIME, and Grad-CAM) were applied to better understand model decisions. The study demonstrates the potential of NIRS combined with AI as a rapid, reliable, and non-destructive tool for improving food allergy detection and safety.

Escherichia coli O157 (E. coli O157) is a well-known foodborne pathogen that significantly threatens to human health and public safety and causes economic loss globally. Given this severe threat, there is an urgent need to develop accurate and timely detection methods for E. coli O157. In this study, a new method was developed for detecting E. coli O157, which combined recombinase polymerase amplification (RPA) with polymer flocculation sedimentation (PFS). Firstly, four upstream primers and 12 downstream primers were designed, and 12 pairs of primers were combined. Then, the optimal primer combinations were screened by using both PCR and RPA methods. Meanwhile, the RPA reaction conditions and the concentration range of PEG8000/NaCl were optimized. Subsequently, the specificity and sensitivity analyses of RPA-PFS assay were further conducted. Finally, artificially contaminated raw milk samples with E. coli O157 and real raw milk samples used to evaluate the potential application of the RPA-PFS method. The results showed that EO3 (F1/R3) was the best primer combination with the optimal reaction temperature, shortest reaction time, and minimum reaction volume of 39/40 °C, 15 min and 10 μL, respectively. The optimal concentration of PEG8000/NaCl binding solution was (0.25 g/mL)/(2.5 M). RPA-PFS assay was specific for 11 non-target foodborne bacteria and could detect E. coli O157 genomic DNA at a detection limit of 2.7 fg. Furthermore, RPA-PFS assay could successfully detect E. coli O157: H7 with 43 CFU/mL in artificially contaminated raw milk within 20 min and was negative for real raw milk sample regardless of bacteria enrichment. In summary, RPA-PFS assay established in this study is a rapid, sensitive, specific, and visual detection tool for E. coli O157, and might be used in resource-limited areas.