Extraction Method Research Articles

Extraction of mucilage polysaccharides from chia seed by hydrophobic deep eutectic solvents-based three-phase partitioning system: A phase behavior-driven approach

By incorporating the hydrophobic deep eutectic solvents (DESs) into the three-phase partitioning (TPP) technique, a TPP-based method was developed to extract the chia seed polysaccharide (CSP) from chia seed. Through a single-factor experiment and response-surface model, the optimal condition for the TPP extraction was determined as DES composed of dodecanoic acid and octanoic acid in a 1:1 M ratio, (NH4)2SO4 concentration of 32.86 %, crude extract–DES ratio of 0.93 (v/v), aqueous phase pH of 4.38, extraction temperature of 35 °C, and extraction time of 10 min. The polysaccharide yield of the constructed TPP method is 8.65 %, which is higher than the conventional water extraction method (yield is 6.96 %). Molecular dynamics simulations reveal the phase behavior of proteins and polysaccharides in the TPP system, showing that noncovalent interactions play a crucial role in the TPP system. The CSP obtained by the TPP method exhibits distinctive composition, structural, physicochemical, and functional properties, leading to improved thermal stability, rheological behavior, and antioxidant performance. Compared with the traditional extraction method, efficient extraction of CSP can be achieved flexibly using the proposed TPP approach, resulting in high yield and quality of CSP, which provides a new path for the large-scale utilization of chia seed.

Using pack animals instead of tractors in Central Italy's protected areas: No evidence of reduced soil disturbance

In thinning operations in Mediterranean beech forests, pack animals are often used when topographic conditions do not allow for the use of mechanisation to perform logging operations. Otherwise, mechanised extraction is the primary choice. In an effort to reduce soil disturbance caused by mechanised ground-based logging, the forest authorities who guard protected areas have been mandating animal logging at all logging sites in recent years. Since no comparative scientific trials of soil disturbance of these two extraction systems have ever been conducted in the target setting, there is no scientific evidence to sustain this approach. On this basis, we chose two study areas where the forest authorities mandated animal logging. Each study area consists of three sub-compartments, one unharvested, one harvested by animals, and one harvested by tractor in the same year (2021). We examined soil compaction, soil organic matter, and soil microarthropod biodiversity in animal skid trails, tractor skid trails, soil unaffected by extraction methods, and unharvested control sites in each study areas. We used linear mixed-effects models, indicating the study area as random intercept, to investigate the effect of the experimental treatments on the target soil features, and to check the relationship among soil physico-chemical parameters and soil microarthropod biodiversity. We discovered that only the animal and tractor skid trails showed signs of disturbance in terms of soil compaction and organic matter loss. Additionally, there was a significant decline in the biodiversity of soil microarthropods in both the skid trail types, with values of the QBS-ar index which were practically halved in comparison to the control area. We did not find any significant difference in the values in the tractor and animal skid trails for any of the examined soil characteristics. Low soil microarthropod biodiversity was found to be correlated with high soil compaction in terms of penetration resistance and low soil organic matter levels, according to linear mixed-effects models. Despite our preliminary findings, we made it clear that the short wood system by a tractor fitted with forwarding bins is unquestionably a very soil-disturbing extraction system, that ought to be avoided when conducting logging operations in protected areas. However, pack animals are not a suitable solution to reduce soil disturbance. Other extraction systems such as chute, small-scale cable yarders, as well as combination of animals and machine avoiding extensive traffic on the forest soil, are better solutions to the fundamental issue of low-impact logging in protected areas.

Determination of 42 antimicrobials in disinfection products by dispersive solid phase extraction-high performance liquid chromatography-tandem mass spectrometry

Antimicrobials inhibit the growth and reproduction of microorganisms thereby alleviating skin and other problems caused by microorganisms. Antimicrobials are classified into different categories, including antibacterials, antifungals, and antivirals, among others, and include azoles, sulfonamides, tetracyclines, quinolones, and many other classes of synthetic and natural compound. The inappropriate or excessive use of antimicrobials can damage skin and other human organs and increase antimicrobial resistance. Relevant regulations and standards clearly state that antimicrobials are prohibited for use as ingredients in disinfection products. However, since antimicrobials enhance the disinfection or antibacterial effect of a product, with a significant short-term effect, antimicrobials are occasionally illegally added to disinfectant products, including those intended for human use. Therefore, establishing testing methods that provide technical support for enforcing regulations is an urgent objective. Herein, a method was established for the analysis of 42 antimicrobials in disinfection products, that is applicable to common types of disinfection-product matrix, including creams, gels, and aqueous solutions, using high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) combined with dispersive solid phase extraction. The 42 antimicrobials comprise antibacterials, antifungals, and antivirals, and include seven sulfonamides, ten quinolones, three lincosamides, five tetracyclines, three macrolides, eight azoles, three purine nucleoside analogs, one furan, one nonpolyene antifungal, and one steroid. Briefly, 0.2 g of a sample was first dispersed in 2 mL of water and then extracted with 10 mL of 0.5% formic acid in acetonitrile, with 3 g of anhydrous Na2SO4 added to remove water. After centrifugation, 5 mL of the supernatant was cleaned using dispersive solid phase extraction with EMR-Lipid as the adsorbent. Lipids, waxes, surfactants, and moisturizing lubricants are commonly used as cream and gel matrices. Matrix substances containing long carbon chains dissolved in acetonitrile were removed using the EMR-Lipid adsorbent. Target analytes were separated on a Poroshell 120 EC-C18 analytical chromatography column (150 mm×3.0 mm, 2.7 μm), with 0.1% formic acid in acetonitrile and 0.1% formic acid aqueous solution used as mobile phases under gradient-elution conditions. The target analytes in the test solution were detected in positive ionization (ESI+) and multi-reaction-monitoring (MRM) modes. Analytes were characterized in terms of their retention times and selected ions, and quantified using the external-standard method. The main factors affecting method response, recovery, and sensitivity, such as the extraction method and solvent, purification method and adsorbent, mobile phase, and MS conditions, were examined during sample pretreatment and instrumental analysis. The 42 antimicrobials were effectively separated under the optimized experimental conditions; the target compounds exhibited linear working curves in the 0.25-5.0 mg/kg concentration range, with correlation coefficients (r) greater than 0.99. Limits of detection (LODs) for the 42 antimicrobials were determined from the signal-to-noise ratios (S/N) of their chromatographic peaks. LODs of 0.03-0.10 mg/kg were determined for the three matrices using 0.2-g samples and 10-mL test solution. Recoveries of 80.3-109.8%, with relative standard deviations (RSDs) of less than 9.8%, were obtained by determining three levels of each target analyte added to the three blank matrices; this process was repeated for six parallel samples. The developed method was used to analyze antimicrobials in commercially available disinfection products, with two sample batches testing positive. The established method is simple, accurate, precise, and suitable for the rapid screening and quantification of antimicrobials in disinfection products. This study provides powerful technical support for regulating the illegal addition of related antimicrobials to disinfection products.

Designing flexible TiO2/SrTiO3/BiOBr recyclable fiber membrane with spatial dual oxidation sites for photocatalytic uranium extraction under visible light

The utilization of nuclear energy generates a substantial volume of wastewater containing uranium, posing a significant threat to both aquatic ecosystems and human health. The current photocatalytic materials utilized for uranium extraction commonly exhibit insufficient redox capabilities, inefficient separation of photogenerated carriers, and challenging recovery. Therefore, recyclable TiO2/SrTiO3/BiOBr (TSB) dual Z-scheme heterojunction fiber membrane materials with dual oxidation sites were successfully synthesized, with advantages of rapid carrier separation and transfer and a broad photo-response range of the dual Z-scheme heterojunctions. Due to the presence of dual oxidation sites, TSB exhibited an exceptional uranium extraction performance, achieving a remarkable 98.9 % extraction rate without the need for sacrificial agents, which was 2.48, 1.79, and 3.34 times of those of ordinary TiO2, SrTiO3, and BiOBr, respectively. The extraction rate of uranium from TSB was 85 % after five cycles. Optical and photoelectric tests revealed that TSB reduced the bandgap width by 0.25 eV compared to that of TiO2, enhanced the fluorescence lifetime by 1.3 ns, lowered the electrical resistance, and higher optical current density, confirming the effective charge transport at the catalyst interfaces. The charge transfer routes in the TSB dual Z-scheme were confirmed by X-ray photoelectron spectroscopy and density functional theory. The potent oxidizing ·OH generated by TSB played a crucial role in facilitating the conversion of U(VI) into insoluble (UO2)O2·2H2O. In conclusion, this work provides a novel strategy for the design of dual Z-scheme heterojunctions and offers a new method for practical and recoverable uranium extraction.

A fast deployable model for crack identification with laser thermography testing

This paper presents a novel feature extraction method, enabling efficient training and deployment of neural networks for rapid identification of crack defects in Laser Array Spot Thermography (LAST). We trained the crack defect identification model based on pixel-level features, encoding each pixel as a feature vector using Frangi filter, and classifying them using a neural network. Experimental results demonstrate that Frangi features are an effective method for distinguishing cracks, speckles, and background noise interference in the experiment. Furthermore, the model only requires a small region of interest (ROI) as training samples to achieve effective training and efficient crack identification under the same detection conditions, allowing for rapid deployment in practical inspections.

Effects Of Extraction Methods on The Physicochemical Characteristics, Fatty Acid Composition, Total Phenol Content, and Alpha-Tocopherol Levels of Moringa Peregrine Seed Oil

Effects Of Extraction Methods on The Physicochemical Characteristics, Fatty Acid Composition, Total Phenol Content, and Alpha-Tocopherol Levels of Moringa Peregrine Seed Oil

Identification of apple watercore based on ConvNeXt and Vis/NIR spectra

This paper proposes a method for discriminating between normal apples and watercore apples using visible/near-infrared spectroscopy technology, which combines Gramian Angular Fields (GAF) encoding technology and the ConvNeXt deep learning network. The existing feature extraction methods for visible/near-infrared spectroscopy data do not perform deeper information mining on the extracted features, which results in the quality of the established model being entirely determined by the extracted features. Additionally, the process of building a visible/near-infrared spectroscopy data classification model is complex and time-consuming, and the accuracy of the established model is not high. To address these issues, the experimental visible/near-infrared spectroscopy data of apples was first transformed into two-dimensional images using Gramian Angular Summation Fields (GASF) and Gramian Angular Difference Fields (GADF) with sizes of 64, 128, 256, and 512. These images were then input into the ConvNeXt network, and the performance of different encoding methods and sizes was compared. The results showed that, under the conditions provided in this paper, the GADF encoding method with a size of 256 achieved the highest classification accuracy of 98.48%. Next, ResNet, EfficientNet, and RegNet deep learning networks were selected to classify the encoded images under the same conditions. The results above indicate that the apple variety discrimination method based on GAF encoding technology and ConvNeXt network combined with visible/near-infrared spectroscopy technology can achieve deep information mining of visible/near-infrared spectroscopy data and provide a relatively simple method for establishing qualitative classification models of visible/near-infrared spectroscopy. This method has a relatively excellent discrimination effect between normal apples and watercore apples.

Review of methods for lithium extraction from geothermal brines

Lithium, a highly reactive and valuable metal, is essential for the clean energy transition, powering electronic devices, electric cars, and energy storage systems. With demand for lithium surging, environmentally responsible and economically viable extraction methods are crucial. Traditional sources include brines and mineral clays, but lithium-ion batteries have become a significant secondary source due to their high consumption of lithium. This review explores various extraction methods from geothermal brines, focusing on conventional techniques like solar evaporation, precipitation, and solvent extraction, highlighting their efficiency and limitations. Advanced electrochemical methods are also discussed, including the use of electrochemical ion pumping and electrodialysis, showcasing their potential for high-purity lithium recovery. Direct Lithium Extraction (DLE) technology, which offers over 90% recovery and reduces impurities by over 99%, is identified as a promising approach. The review underscores the need for large-scale field experiments and the development of new lithium sources to meet growing demand.

Optimization of extraction for efficient recovery of kenaf seed protein isolates: evaluation of physicochemical and techno-functional characteristics.

Kenaf seeds are a rich source of protein; however, finding the best extraction method is crucial to obtaining high-quality protein from these underutilized seeds. This research devised an optimized extraction process for best recovery of kenaf seeds protein using response surface methodology. The key parameters affecting the yield and protein content were optimized, including extraction pH (2-11), seed:water ratio (5:1-50:1), temperature (30-90 °C), and duration (20-360 min). The physicochemical and techno-functional properties of kenaf seed protein isolates (KSPIs) were examined. A maximum protein yield of 12.05 g/100 g with purity level 91.94 g/100 g was obtained using an optimized extraction with pH 11.0, seed:water ratio 50:1, 360 min duration, and temperature 50 °C. The oil and water retention capacities of KSPI were 1.14 mL g-1 and 1.37 mL g-1 respectively. After 30 min at pH 7, KSPIs demonstrated remarkable emulsion capacity (83.12%) and stability (75.63%), along with high foaming capacity (106%) and stability (18.3%). As per high-performance liquid chromatography analysis, arginine, glutamic acid, leucine, phenylalanine, and lysine were the most abundant amino acids detected in KPSIs. The KSPIs' globular protein structure was successfully verified using analytical approaches, including Fourier transform infrared spectroscopy, protein fraction ratios, and differential scanning calorimetry. Sodium dodecyl sulfate polyacrylamide gel electrophoresis analysis revealed that KPSI has a molecular weight distribution ranging from 10 kDa to 50 kDa. The results of this study support the application of the proposed response-surface-methodology-optimized extraction method for efficient recovery of high-quality kenaf seed proteins that meet the necessary physicochemical and techno-functional requirements. © 2024 Society of Chemical Industry.

Drug–drug interaction extraction based on multimodal feature fusion by Transformer and BiGRU

Understanding drug–drug interactions (DDIs) plays a vital role in the fields of drug disease treatment, drug development, preventing medical error, and controlling health care-costs. Extracting potential from biomedical corpora is a major complement of existing DDIs. Most existing DDI extraction (DDIE) methods do not consider the graph and structure of drug molecules, which can improve the performance of DDIE. Considering the different advantages of bi-directional gated recurrent units (BiGRU), Transformer, and attention mechanisms in DDIE tasks, a multimodal feature fusion model combining BiGRU and Transformer (BiGGT) is here constructed for DDIE. In BiGGT, the vector embeddings of medical corpora, drug molecule topology graphs, and structure are conducted by Word2vec, Mol2vec, and GCN, respectively. BiGRU and multi-head self-attention (MHSA) are integrated into Transformer to extract the local–global contextual DDIE features, which is important for DDIE. The extensive experiment results on the DDIExtraction 2013 shared task dataset show that the BiGGT-based DDIE method outperforms state-of-the-art DDIE approaches with a precision of 78.22%. BiGGT expands the application of multimodal deep learning in the field of multimodal DDIE.

Pressurized liquid extraction of sulfite for quantitation by capillary zone electrophoresis.

Sulfite is an additive used in shrimp processing to prevent discoloration. However, sulfite can cause health issues for sensitive consumers, making its monitoring necessary. Determining sulfite concentrations is complex because the Monier-Williams reference method is laborious and has low analytical throughput. Additionally, new techniques are needed to stabilize sulfite during the extraction process because this analyte undergoes rapid oxidation. A new method involving extraction and derivatization of sulfite with formaldehyde through automated pressurized liquid extraction (PLE), followed by quantitation by capillary zone electrophoresis with diode-array detector in indirect mode, was developed and optimized using multivariate planning. The PLE procedure was compared to another solid-liquid extraction method. The new method successfully stabilized and extracted sulfite from shrimp in few steps with adequate precision (CV<3.8%), producing extracts that were stable for 10 days. Recovery was satisfactory (97%-99%), and the limits of detection (4.6mgkg-1) and quantitation (15.4mgkg-1) were suitable for the intended purpose.

Improved Early-Stage Maize Row Detection Using Unmanned Aerial Vehicle Imagery

Monitoring row centerlines during early growth stages is essential for effective production management. However, detection becomes more challenging due to weed interference and crop row intersection in images. This study proposed an enhanced Region of Interest (ROI)-based approach for detecting early-stage maize rows. It integrated a modified green vegetation index with a dual-threshold algorithm for background segmentation. The median filtering algorithm was also selected to effectively remove most noise points. Next, an improved ROI-based feature point extraction method was used to eliminate residual noises and extract feature points. Finally, the least square method was employed to fit the row centerlines. The detection accuracy of the proposed method was evaluated using the unmanned aerial vehicle (UAV) image data set containing both regular and intersecting crop rows. The average detection accuracy of the proposed approach was between 0.456° and 0.789° (the angle between the fitted centerline and the expert line), depending on whether crop rows were regular/intersecting. Compared to the Hough Transform (HT) algorithm, the results demonstrated that the proposed method achieved higher accuracy and robustness in detecting regular and intersecting crop rows. The proposed method in this study is helpful for refined agricultural management such as fertilization and irrigation. Additionally, it can detect the missing-seedling regions and replenish seedings in time to increase crop yields.

Cinnamon – A Competent Drug: A Review on Extraction, Analysis and Anticancer Action

AbstractCinnamon, an Indigenous species, is extensively used as a folk medicine in India, China, and other parts of the world due to its therapeutic potential inherited via the latent chemical composition. The vital component presented is cinnamaldehyde, along with cinnamic acid and cinnamate, which contributes to being an anti‐inflammatory, antimicrobial, antidiabetic, and anticancer agent together with the capability to control neurological syndromes like Alzheimer's and even Parkinson's diseases. Given the importance of the anticarcinogenic properties of cinnamon on various cell strains concerning the curable effect, this review focuses on evaluating different extraction methods like steam distillation, Soxhlet extraction, microwave‐assisted extraction, and more, in addition to a summary of new technologies like gas chromatography, HPLC, DART‐MS, and NMR, etc. which paved the way in characterizing the chemical composition of cinnamon. Cinnamaldehyde showed its apoptosis through various mechanistic pathways on an adequate number of cell lines and antineoplastic potential on specific multifaceted cancerous cells, which advocates for continued research and investment in this vital area of drug discovery and suggestions for future scope.

Optimizing the extraction of essential oil yield from Pistacia lentiscus oleo-gum resin by superheated steam extraction using response surface methodology

Pistacia lentiscus L. is an aromatic plant containing a significant percentage of essential oil (EO) used in fragrance, pharmaceuticals, cosmetics, and the food industry. The purpose of this work is focused on the optimization of Pistacia lentiscus L. oleo gum resin EO yield extracted by superheated steam extraction (SHSE) by response surface methodology, including extraction parameters of particle size (0. 5 − 1 mm), temperature (140–180 °C) and time (90–150 min). The optimum conditions for Pistacia lentiscus L. EO extracted by SHSE were found to be (particle size: 0.75 mm, time: 120 min and temperature: 160 ℃) which produced the highest EO yield of 5.7%. A regression model was developed, demonstrating a robust quadratic correlation with an R2 value of 0.9991, making it suitable for predictions. Furthermore, the yield of Pistacia lentiscus L. EO extracted by SHSE was compared with the conventional steam and hydro distillation techniques. The study revealed that SHSE yielded higher quantities of EO than other extraction methods. GC-MS analyzed the chemical composition of Pistacia lentiscus L. EO. The predominant compound of Pistacia lentiscus L. EO was determined to be α-pinene, while the other identified compounds include trans-verbenol, verbenol, cis-verbenone, camphene, β-myrcene, d-limonene, cymene, α-myrtenol, α-campholenal, α-copaene, and α-thujene, whose content differed according to different extraction techniques. Overall, superheated steam extraction is an efficient technique for extracting Pistacia lentiscus L. essential oil that enhances EO yield, requiring less time for extraction.

Anchoring temporal convolutional networks for epileptic seizure prediction.

Accurate and timely prediction of epileptic seizures is crucial for empowering patients to mitigate their impact or prevent them altogether. Current studies predominantly focus on short-term seizure predictions, which causes the prediction time to be shorter than the onset of antiepileptic, thus failing to prevent seizures. However, longer epilepsy prediction faces the problem that as the preictal period lengthens, it increasingly resembles the interictal period, complicating differentiation. To address these issues, we employ the sample entropy method for feature extraction from electroencephalography (EEG) signals. Subsequently, we introduce the Anchoring Temporal Convolutional Networks (ATCN) model for longer-term, patient-specific epilepsy prediction. ATCN utilizes dilated causal convolutional networks to learn time-dependent features from previous data, capturing temporal causal correlations within and between samples. Additionally, the model also incorporates anchoring data to enhance the performance of epilepsy prediction further. Finally, we proposed a multilayer sliding window prediction algorithm for seizure alarms. Evaluation on the Freiburg intracranial EEG dataset shows our approach achieves 100% sensitivity, a false prediction rate (FPR) of 0.08 per hour and an average prediction time (APT) of 99.98 minutes. Using the CHB-MIT scalp EEG dataset, we a achieve 97.44% sensitivity, an FPR of 0.11 per hour, and an APT of 92.99 minutes. These results demonstrate that our approach is adequate for seizure prediction over a more extended prediction range on intracranial and scalp EEG datasets. The average prediction time of our approach exceeds the typical onset time of antiepileptic. This approach is particularly beneficial for patients who need to take medication at regular intervals, as they may only need to take their medication when our method issues an alarm. This capability has the potential to prevent seizures, which will greatly improving patients' quality of life.

Insights into type 2 diabetes mellitus in atrial fibrillation patients: a proteomics approach

Abstract Introduction Atrial fibrillation (AF) and Type 2 Diabetes Mellitus (T2DM) are two closely interconnected conditions, sharing a complex relationship within cardiac pathophysiology. Common risk factors such as obesity, hypertension, inflammation, and endothelial dysfunction contribute to their coexistence. T2DM is known to drive cardiac fibrosis and promote electrical and structural remodeling of the heart, creating a substrate conducive to AF onset, while AF in T2DM patients increases the risk of adverse cardiovascular outcomes, such as stroke. Understanding this complex relationship is pivotal for improving the therapeutic and preventive strategies for these patients. This study aims to examine the proteomic differences between AF patients with and without T2DM, providing novel insights into pathophysiological mechanisms and identifying novel therapeutic targets and biomarkers. Methods In total, 14 AF patients undergoing open-heart surgery, comprising of 6 T2DM individuals and 8 non-T2DM individuals have been included. The mean age of the T2DM patients was 69.5 ± 10 years, while the non-T2DM was 69.3 ± 8.2 and BMI was 29.2 ± 2.9 and 29.5 ± 3 respectively. Right atrial appendages (RAA) from these patients were obtained in an RNA stabilization solution during open-heart surgery and processed using the phenol-based method for protein extraction. The proteins were digested with trypsin to generate peptides, followed by an additional clean-up step using C18 pipette tips. The prepared peptide samples were analyzed by employing a label-free-based quantitative (LFQ) approach, using a C18 (75cmx75cm) column. Proteomic analysis was performed using a Nano liquid chromatography system coupled to tandem mass spectrometer (nano LC-MS/MS), followed by a comprehensive bioinformatics analysis, using appropriate software. Results A total of 1548 proteins were identified, of which 899 could be quantified using LFQ analysis. Applying strict significance criteria, including FDR &amp;lt;0.1 and Abundance ratio (AR)&amp;gt;2 or &amp;lt;0.5, we identified 100 differential expressed proteins, with 21 upregulated and 79 downregulated in T2DM (Fig.1). Most notably, DPP9 and ATP1A2 were overexpressed in hearts of AF patients with T2DM (AR=100), while INPPL1 and ARID5A were underexpressed (AR=0.01), thereby representing attractive therapeutic targets. In pathway enrichment analysis (Fig.2), the dysregulated proteins participate in noteworthy pathways, including Non-alcoholic fatty liver disease (NAFLD), Type II diabetes mellitus, and Insulin signaling pathway. Conclusion Given the increased morbidity associated with the coexistence of AF and T2DM, there is a pressing need for novel therapeutic strategies tailored to this patient population. While AF and T2DM share certain molecular characteristics in the heart, this study emphasizes the necessity for further evaluation and validation of the distinct proteomic signatures identified.Fig.1:Volcano PlotFig.2:Protein Enrichment Network

Mercury speciation in environmental samples associated with artisanal small-scale gold mines using a novel solid-phase extraction approach to sample collection and preservation

In artisanal small-scale gold mines (ASGM), mercury (Hg) is known to pollute nearby river waters and sediments where it can be methylated to the highly bioavailable methylmercury (MeHg). The assessment of Hg speciation in water samples has been challenging for many years, with recommended procedures often not adequately allowing for analysis of samples in a suitable timeframe. Using a novel solid-phase extraction (SPE) method for sampling and preservation of Hg species, representative speciation data can be safely and easily collected and retained for up to 4-weeks (MeHg = 115 ± 8% refrigerated and 109 ± 13% unrefrigerated storage; Hg2+ = 100 ± 14% refrigerated and 94 ± 12% unrefrigerated storage). Concentrations of MeHg in environmental water samples and drinking water were below detection limit across two ASGM sites in western Kenya and concentrations of Hg2+ were below drinking water guidelines; however, drinking water sources contribute 20–30% of the tolerable weekly intake of Hg, indicating a need to minimise exposure of Hg from dietary sources to prevent Hg poisoning. Sediments from receiving rivers at ASGM sites showed total Hg concentrations above guideline limits (0.08–1.84 mg kg−1 total Hg) along the length of the river; however, MeHg concentrations fluctuated dependent on the stagnation of the river due to damns and ponds (5.9 ± 14.3 µg kg−1 MeHg). The findings show that SPE can be used as a robust sample collection and preservation approach for Hg speciation, which can better inform mitigation measures, understand ecological and human health implications, and improve environmental monitoring.

A covalent organic framework-derived pretreatment for pesticides in vegetables and fruits

Sample pretreatment is an essential procedure in pesticide analysis, as the matrix effect can significantly influence the results. In this study, a covalent organic framework (COF) was synthesized using 1,2,4,5-tetrakis-(4-formylphenyl)benzene (TFPB) and benzidine (BD) to mitigate the matrix effect in vegetable and fruit samples. This COF was then used to develop a solid-phase extraction (CSPE) method. In addition, the COF was used to create a magnetic COF (MCOF) for use in magnetic solid-phase extraction (MSPE). The synthesized COF and MCOF were thoroughly characterized using scanning electron microscopy (SEM) for morphological analysis, Fourier-transform infrared spectroscopy (FT-IR) for chemical bond identification, and N2 adsorption–desorption measurements for porosity and surface area assessment. Key pretreatment parameters such as buffers, dilution rate, sorbent dosage, extraction time, elution solvent, and reuse number were optimized. The developed CSPE and MSPE showed excellent purification ability for the matrix of vegetable and fruit samples. The reuse test demonstrated that the synthesized COF and MCOF can be reused up to 15 times. Moreover, the developed CSPE and MSPE showed acceptable recoveries in spiked recovery tests, suggesting that these pretreatment methods were feasible for sample purification in pesticide analysis.

Identification and Analysis of Phenolic Compounds in Vaccinium uliginosum L. and Its Lipid-Lowering Activity In Vitro

Vaccinium uliginosum L. (VU), rich in polyphenols, is an important wild berry resource primarily distributed in extremely cold regions. However, the detailed composition of Vaccinium uliginosum L. polyphenols (VUPs) has not been reported, which limits the development and utilization of VU. In this study, VU-free polyphenols (VUFPs) and VU-bound polyphenols (VUBPs) were, respectively, extracted using an ultrasonic, complex enzyme and alkali extraction method; the compositions were identified using ultra-performance liquid chromatography–electrospray ionization mass spectrometry, and lipid-lowering activity in vitro was evaluated. The results showed that 885 polyphenols and 47 anthocyanins were detected in the VUFPs and VUBPs, and 30 anthocyanin monomers were firstly detected in VU. Compared with the model group, the accumulation of lipid droplets and the total cholesterol and triglyceride contents in the high-concentration VUP group reduced by 36.95%, 65.82%, and 62.43%, respectively, and liver damage was also alleviated. It was also found that VUP can regulate the level of Asialoglycoprotein receptor 1, a new target for lipid lowering. In summary, this study provides a detailed report on VUP for the first time, confirming that VUP has lipid-lowering potential in vitro. These findings suggest new strategies and theoretical support for the development and utilization of VU, especially in the field of functional foods.

An inherently interpretable deep learning model for local explanations using visual concepts.

Over the past decade, deep learning has become the leading approach for various computer vision tasks and decision support systems. However, the opaque nature of deep learning models raises significant concerns about their fairness, reliability, and the underlying inferences they make. Many existing methods attempt to approximate the relationship between low-level input features and outcomes. However, humans tend to understand and reason based on high-level concepts rather than low-level input features. To bridge this gap, several concept-based interpretable methods have been developed. Most of these methods compute the importance of each discovered concept for a specific class. However, they often fail to provide local explanations. Additionally, these approaches typically rely on labeled concepts or learn directly from datasets, leading to the extraction of irrelevant concepts. They also tend to overlook the potential of these concepts to interpret model predictions effectively. This research proposes a two-stream model called the Cross-Attentional Fast/Slow Thinking Network (CA-SoftNet) to address these issues. The model is inspired by dual-process theory and integrates two key components: a shallow convolutional neural network (sCNN) as System-I for rapid, implicit pattern recognition and a cross-attentional concept memory network as System-II for transparent, controllable, and logical reasoning. Our evaluation across diverse datasets demonstrates the model's competitive accuracy, achieving 85.6%, 83.7%, 93.6%, and 90.3% on CUB 200-2011, Stanford Cars, ISIC 2016, and ISIC 2017, respectively. This performance outperforms existing interpretable models and is comparable to non-interpretable counterparts. Furthermore, our novel concept extraction method facilitates identifying and selecting salient concepts. These concepts are then used to generate concept-based local explanations that align with human thinking. Additionally, the model's ability to share similar concepts across distinct classes, such as in fine-grained classification, enhances its scalability for large datasets. This feature also induces human-like cognition and reasoning within the proposed framework.