Agriculture dominates United Kingdom ammonia emissions, from livestock manure exposed to the atmosphere via livestock housing, storage, land and grazing. Ammonia significantly contributes to the formation of PM2.5 (particles with diameter of 2.5 μm or less) concentrations in Europe which are associated with adverse human health outcomes. Ammonia emissions contribute to nitrogen deposition, whereby reactive compounds of nitrogen are deposited into the biosphere, potentially resulting in biodiversity loss. Recent research has not found sufficient evidence for effectiveness of interventions to reduce ammonia emissions and little evidence on the cost-effectiveness of interventions. The current study aimed to address this knowledge gap. The study aimed to assess effectiveness and cost-effectiveness of two agricultural interventions to mitigate ammonia emissions - improved housing for farmed animals and improved manure application. Emission measurements were made at five farms (dairy, pig, poultry). Information on uptake of mitigation measures, barriers and enablers for implementation were determined through an online survey and focus groups with farmers, supplemented by stakeholder interviews. Chemical transport and dispersion modelling estimated population exposures to air pollution at local and national levels under three scenarios (low, medium, high intervention uptake). A health impact assessment estimated health effects associated with the scenarios, and data on self-reported health issues were collected via an online survey of rural residents. Economic evaluation methods estimated cost-benefits of the scenarios and impact on ecosystems. Farmers favour mitigation measures which are cheaper, and build on existing practices, such as amending diet or extending the grazing season. However, these are less effective in decreasing ammonia emissions. Scenarios based on realistic current, and future, uptake levels of measures showed little impact on air quality, partly due to the ammonia-rich United Kingdom atmosphere minimising conversion of ammonia emissions to particulate matter. Consequently, minimal impact of mitigation measures was evident on health outcomes and costs. There was no evidence that self-reported health symptoms from rural residents were related to living near a farm, type of farm or seasonality of farm activities, consistent with results of local dispersion modelling which estimated that most emissions from animal housing dispersed within 1 km. Impacts of COVID-19 and the United Kingdom's withdrawal from the European Union on the agricultural industry affected the recruitment and availability of farms and farmers, resulting in fewer field measurements than planned. A lower response to the farmers' survey was mitigated by the quality of data provided by participants and the successful series of focus groups. The study highlights the need for enhanced communication with the farming community to encourage implementation of more effective mitigation measures, such as air scrubbers, or those relating to slurry storage, currently perceived to be too expensive and complex. Greater clarity on benefits is essential so that farmers understand not only what they need to do but also how and why. Further investigation of the health impacts of ammonia emission should focus on those exposed on the farm, or resident nearby animal houses. Further modelling development of key atmospheric processes is also indicated to minimise uncertainties associated with the regional modelling. This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme as award number NIHR129449. Air pollution damages lung and heart health, contributing to premature death and hospital admissions. These health effects are associated with exposure to very small particles, including from reactions of ammonia emitted from farming, the main United Kingdom producer of ammonia, principally from animal manure in livestock housing, fertilisation of fields and animal grazing. Recently, other United Kingdom sources of particles have decreased, but ammonia levels have not. This study aimed to assess the effectiveness of improved cattle housing and manure storage and application, at reducing emissions of gases and particles. We measured ammonia emissions from five farms and used surveys, focus groups and interviews with farmers and stakeholders to understand views on ammonia reduction measures. Computer models were used to estimate the impact of emissions reduction on exposure and related health issues of people near farms and the wider United Kingdom population. We calculated savings in National Health Service costs. We also surveyed people living near farms about their health. The study found that the measures that farmers were currently prepared to consider implementing reduced ammonia emissions by up to 13%, but the overall reduction in air pollution particles was limited (around 1%). Improvements in health and cost-savings were also small, and surveys of rural residents did not show health problems were related to farming. The study also showed that emissions from farms almost entirely dispersed into the background air within 1 km. Farmers were interested in reducing their environmental impact and favoured cheaper interventions building on existing practices, which also tended to be less effective in reducing ammonia emissions. Barriers to using these interventions were costs and lack of knowledge. To reduce ammonia emissions, future policies should address the barriers and clearly communicate benefits to the environment and to farmers. It would be useful to study more effective farming interventions to reduce air pollution.
BackgroundLarge language models (LLMs) may reduce the burden associated with performing systematic reviews by prescreening abstracts from a literature search for eligibility for inclusion in full-text review.MethodsWe developed an iterative, LLM-based workflow for screening abstracts: after manual specification of eligibility criteria and seed examples, an ensemble of five LLMs deliberates through a Delphi process to classify a batch of abstracts; these labels are used to train a logistic regression model that ranks the remaining abstracts and identifies a new batch of abstracts for LLM escalation until all abstracts are labelled by the LLM or probability thresholds. We tested our workflow on abstracts screened in three published systematic reviews in psychiatry. Our primary endpoint was the recall metric, and secondary endpoint was the work saved over sampling at 95% recall metric (WSS@95%).ResultsIn a dataset on autism biomarkers, 1,655 (35%) of 4,745 retrieved abstracts were judged to be relevant by the original authors. The Delphi-LLM workflow correctly identified 1,605 (97.0%) of these 1,655 abstracts (precision = 54.2%, WSS@95% = 38.1%). The performance metrics were better than non-LLM approaches (recall ≤ 91%, WSS@95 ≤ 26%), and, overall, balanced these metrics optimally compared to single-LLM agents (recall = 84.9-99.9%, WSS@95% = 16.7-39.8%). The recall and work saved metrics were similarly reliable and among the top in two low-prevalence datasets on an attention-deficit hyperactivity disorder treatment review (10% of 2,891 relevant) and a posttraumatic stress disorder trajectory review (7% of 4,453 relevant). For these two datasets, recall was 100.0% and 96.4%, and the WSS@95% was 17.3% and 18.5%, respectively.ConclusionsWe presented the design and validation of a novel abstract screening workflow that centres around a Delphi-style aggregation process to harness the strengths of five open-source LLMs that can be run on consumer-level workstations. This multi-LLM workflow showed acceptable and reliable performance for use as an automated prescreening method to facilitate systematic reviews. A new algorithm to debate a team of AI language models for improving the screening process of article abstracts in psychiatry.Plain Language SummaryLarge language models (LLMs) can now understand and write complex scientific texts. Systematic reviews, which demand significant time and resources, can greatly benefit from the use of LLMs, which can process thousands of articles much faster and cheaper than human researchers.We built an automated workflow inspired by the Delphi method, where five different LLMs communicate with each other to reach a consensus decision on the relevance of a given abstract for a specific systematic review. In essence, the five-LLM Delphi ensemble serves as a teacher model that processes a small batch of difficult abstracts, which are also used to train a student classifier model that can clear similar abstracts. This two-stage cycle repeats until all abstracts are labelled, either by the LLM–Delphi teacher ensemble, or by the student classifier.We tested this workflow on three systematic reviews in psychiatry. In the first review containing 4,745 abstracts, the original authors determined 1,655 (35%) to be truly relevant. The workflow automatically determined 62% of the dataset as relevant and 38% as irrelevant, saving significant labour. The positively labelled abstracts contained 97% of the truly relevant abstracts, demonstrating the workflow’s competence (i.e., recall). The workflow showed similarly high recall (96–100%) but reduced labour savings (17–18%) in the other two reviews, which contained a lower proportion of relevant abstracts (i.e., 7% of 4,453 total in one, and 10% of 2,891 in another). Overall, the five-LLM model produced a more well-rounded and reliably excellent performance compared to using just a single-LLM model as the teacher, and better performance compared to not using LLMs to label the difficult cases.The entire workflow can run on a consumer-grade computer without sending data to costly cloud-based models. By achieving high recall ability and substantial labour savings, this approach could make systematic reviews faster and cheaper to conduct. Les grands modèles de langage (GML) peuvent réduire le fardeau associé à la réalisation de revues systématiques en effectuant une présélection des résumés issus d’une recherche documentaire pour déterminer s’ils peuvent être inclus dans une revue de texte intégral. Nous avons mis au point un flux de travail itératif fondé sur des GML en vue de la sélection des résumés; après avoir précisé manuellement les critères d’admissibilité et fourni des exemples initiaux, un ensemble de cinq GML délibère selon un processus Delphi afin de classer un lot de résumés; ces étiquettes servent ensuite à entraîner un modèle de régression logistique qui classe les résumés restants et identifie un nouveau lot de résumés à soumettre aux GML jusqu’à ce que tous les résumés aient été classés par les GML ou sur la base de seuils de probabilité. Nous avons testé notre flux de travail sur des résumés examinés sélectionnés dans le cadre de trois revues systématiques publiées en psychiatrie. Notre critère d’évaluation principal était la mesure du rappel, et notre critère d’évaluation secondaire était la quantité de travail évité par rapport à l’échantillonnage (work saved over sampling [WSS]) pour atteindre un rappel de 95% (WSS à 95%). Dans un ensemble de données portant sur les biomarqueurs de l’autisme, 1 655 (35%) des 4 745 résumés récupérés ont été jugés pertinents par les auteurs originaux. Le flux de travail Delphi-GML a correctement identifié 1 605 (97,0%) de ces 1 655 (précision = 54,2%, WSS à 95% = 38,1%). Les indicateurs de rendement se sont révélés supérieurs à ceux des approches ne faisant pas appel aux GML (rappel ≤ 91%, WSS à 95% ≤ 26%), tout en étant plus équilibrés que les meilleurs agents utilisant un seul GML (rappel = 84,9 à 99,9%, WSS à 95% = 16,7 à 39,8%). Les paramètres relatifs au rappel et à la quantité de travail évité (work saved) étaient tout aussi fiables et figuraient parmi les meilleurs résultats dans le cadre de deux ensembles de données à faible prévalence, l’un portant sur une revue du traitement du trouble du déficit de l’attention avec hyperactivité (10% des 2 891 résumés étant pertinents) et l’autre sur une revue des trajectoires du trouble de stress post-traumatique (7% des 4 453 résumés étant pertinents). Pour ces deux ensembles de données, le rappel était de 100,0% et de 96,4%, et les WSS à 95% étaient de 17,3% et de 18,5%, respectivement. Nous avons présenté la conception et la validation d’un nouveau flux de travail relatif à la sélection de résumés axé sur un processus d’agrégation de style Delphi et visant à exploiter les atouts de cinq GML à code source ouvert pouvant être utilisés sur des postes de travail grand public. Ce flux de travail multi-GML a montré un rendement acceptable et fiable en vue de son utilisation comme méthode de présélection automatisée pour faciliter les revues systématiques.
Cigars remain among the most affordable tobacco products in the U.S., with package features designed to increase appeal, including pack size, flavor, price, and potentially misleading descriptors, like "natural." Building upon prior work showing frequent changes in the cigar market, we examined trends in US cigar unit sales and prices by package feature from 2021 to 2025. Nielsen Convenience Track national cigar sales data from 2021 to 2025 were used for analyses. Total cigar unit sales decreased from $2.22 billion in 2021 to $1.79 billion in 2025, while average price per stick increased from $2.86 to $3.91. In 2025, the pack sizes with the greatest market share were two cigars/pack (46.8% of market) and single sticks (30.3%). In 2025, over one-quarter of cigar sales featured a "natural" descriptor; unflavored cigars had the greatest market share (43.3%), followed by flavored cigars (34.0%), and cigars with concept (i.e., ambiguous) descriptors (22.7%). From 2021 to 2025, market share decreased for flavored and unflavored cigars but increased for concept cigars. For average price per cigar, in 2025, two-packs, three-packs, and 20-packs were the cheapest per cigar compared with other pack sizes ($0.76, $0.75, and $0.66, respectively), with price per cigar for two- and three-packs decreasing from 2021 to 2025. Flavored cigars ($1.03) were cheaper than unflavored ($6.84) and concept cigars ($2.23) per cigar. Cigars with a "natural" descriptor were cheaper ($1.42) than those without ($4.60). Findings reflect a diverse cigar market with respect to pack size, flavor, and marketing descriptors.
The Framework Convention on Tobacco Control includes tobacco price promotion bans as a part of a comprehensive tobacco control strategy. The authors examined how U.S. adults who use commercial tobacco would react to a hypothetical national ban on tobacco price promotions. Data were from a cross-sectional survey of a nationally representative sample of U.S. adults (aged ≥21 years) between March and May 2024. Participants who used discount coupons and price promotions during the 12 months prior to the survey and were currently using commercial tobacco (N=608) reported whether they would engage in a list of behaviors more frequently to save money on tobacco (e.g., buying a cheaper brand, quitting tobacco and E-cigarette use) in the case of a national ban on tobacco discount coupons and price promotions. Weighted prevalence of these behaviors and predicted marginal probabilities by demographics were estimated. Among U.S. adults who used tobacco discount coupons and promotions, 33.1% reported that they would quit using tobacco and E-cigarettes in response to a nationwide ban on tobacco price promotions. However, 95.9% reported that they would engage in tobacco expenditure minimizing strategies more frequently (e.g., finding less expensive places to buy cigarettes or E-cigarettes=73.0%, purchasing by bulk=58.7%, using a cheaper form of tobacco=38.7%). The probabilities of engaging in these strategies varied somewhat by demographics. Although a national ban on tobacco price promotions may promote tobacco use cessation, most adults who used tobacco price promotions would increase tobacco expenditure minimizing strategies use to continue commercial tobacco use, highlighting the importance of simultaneously addressing shifts in tobacco expenditure minimizing strategies use.
The 3ω method has recently been proposed for in situ and operando measurements, for example, lithium concentration detection in lithium-ion pouch cells [e.g., Zeng et al., Joule 5, 2195-2210 (2021)] and human health monitoring [e.g., Qiu et al., Int. J. Heat Mass Transfer. 163, 120550 (2020)]. Yet, the bulky and expensive instruments hinder these proposals from moving toward practical applications. Here, we develop a portable 3ω setup that drives the measurements by a data acquisition card and detects the weak 3ω voltage signal through a homemade digital phase-sensitive (lock-in) detector. This portable setup is at least an order of magnitude lighter, smaller, and cheaper than its conventional laboratory counterpart. We validate this setup by measuring the thermal properties of a bulk SiO2 sample and a SiO2-film-on-Si-substrate sample. We conduct synthetic experiments to demonstrate the feasibility of this setup for in situ and operando lithium concentration detection in lithium-ion pouch cells.
Depression is one of the most common mental disorders, and more than half of individuals with depression have reduced or discontinued antidepressant use due to side effects. Medicinal plants are safer, cheaper, and more readily available than synthetic medications. This study will employ bioinformatics to assess the impact of compounds discovered in plants used in traditional Persian medicine (TPM) on key proteins involved in depressive pathways. Compounds derived from plants used in TPM for the treatment of depression were identified and subsequently evaluated for ADME properties and toxicity. The potential target genes of these compounds, along with genes associated with depression, were identified, and the intersecting genes were selected to construct a protein-protein interaction (PPI) network. Molecular docking analysis of the plant compounds and their key target proteins was conducted, followed by the selection of the most effective compound based on molecular dynamics (MD) simulation. In this study, 23 genes were identified as target genes of the plant compounds, and 582 genes associated with depression were identified, nine of which were shared, including ABCB1, AKT1, CAT, CDH1, CYP2B6, ESR1, ESR2, PPARG, and TRPV1. The PPI network identified ABCB1, AKT1, CDH1, ESR1, and PPARG as key proteins. Among them, ABCB1 and AKT1 exhibited favorable docking energies with carnosic acid. MD simulations further revealed the stability of the ABCB1-carnosic acid and AKT1-carnosic acid complexes. Our findings indicate that carnosic acid may represent a potential therapeutic option for the treatment of depression. However, further in vivo and clinical trials are required to validate these findings.
Expensive dairy products like ghee are at high risk of hazardous adulteration with cheaper fats, and traditional testing approaches such as FTIR device and conventional machine learning are too slow, complex, and lack the interpretability to be useful for practical food safety. To address this, we propose a new deep learning framework which is robust and highly interpretable. Our solution is a multimodal attention-based architecture that integrates optimized machine learning classifiers with advanced deep models, including these phases Phase I: Feature Extraction by help of Machine Learning Models (e.g. SVM, XGBoost, Random Forest), Phase II: Extracted Feature converted to spectroscopy, Phase III: an Attention-Augmented 1D-CNN and a Transformer-based network. Importantly, we introduced a Spectral Attention Map (SAM) mechanism that automatically focuses the chemically relevant FTIR bands, ensuring the completely transparent model's high-accuracy decisions beyond the "black-box". The reliability can be enhanced using a Dynamic Multimodal Fusion scheme that learns to assigns weights to the outputs of all models adaptively. Our system was evaluated on a statistically enriched, custom FTIR dataset with various different attributes of ghee on the basis of which we create the spectroscopy of the respective attributes and achieved a staggering classification accuracy of 99.6% with a 94% feature dimensionality reduction, showing the efficiency and effectiveness of the proposed approach. This work establishes a new benchmark for food authentication using AI, offering a highly accurate, scalable and transparent solution with significant potential to be incorporated into a real-time and portable food safety systems to allow on-site verification.
Cheaper forgeries emerge inevitably for commercial interests when authentic products are developed with great market potential. For example, diacylglycerol-enriched oils are often forged by that containing mainly triacylglycerol, the former health beneficial whereas the latter not. To detect such forgeries, fast, efficient and specific methods are necessary. Therefore, we established a hydrolysis system with Penicillium camemberti lipase to enable diacylglycerol-enriched oils to exhibit a distinct pH response from that of forged oils. This different pH could be visible by pH-sensitive films. Further, we optimized the hydrolysis system for higher efficiency, followed by testing the potential of this film-based method in semi-quantifying diacylglycerol level (about 20, 40, 60, and 80%). This method was then validated by comparing with the results from traditional high performance liquid chromatography-based method. Our findings provide an effective method for detecting forged oils from diacylglycerol-enriched oils in non-laboratory settings, contributing to improving authenticity and quality control.
Clinical Prediction Rules (CPR) such as the New Orleans Criteria (NOC), National Institute for Health and Clinical Excellence (NICE), and Canadian Computed Tomography Head Rules (CCHR) are standard triaging guidelines to determine the need for CT in mTBI. This study evaluates the effectiveness of CEREBO®, a machine learning-powered near infrared spectroscopy (NIRS) device, in triaging mTBI patients as compared to CPR. 376 patients with mTBI from 3 single-center studies at NIMHANS Bengaluru, Civil Hospital, Ahmedabad and AIIMS, Bhopal underwent secondary analysis. The patients underwent CEREBO® scanning, and results were compared to CT using the Diagnostic Efficiency and Reduction of CT scans (D-ERCS) Framework. Data from published literature was used to compare with CPR. The diagnostic accuracy of CEREBO®, reduction in unnecessary CT scans and number-needed-to-scan (NNS) in comparison to CPR were calculated. CEREBO® demonstrated superior diagnostic accuracy - 96.6% sensitivity, 84.2% specificity, reducing unnecessary CT scans by 67.4-85.7% in pediatric, 63-83.8% in adult, and 20.3-65.7% in geriatric populations. CEREBO® achieved an NNS of 3-4 as compared to 9-13 scans with CPR, demonstrating efficiency in identifying positive cases. The usage of CEREBO® was 94%, 95.6% and 95.2% cheaper as compared to the NOC, CCHR and NICE guidelines respectively. CEREBO® is a reliable tool for triaging mTBI, offering accurate diagnoses while significantly reducing unnecessary imaging. By lowering radiation exposure. optimizing resource use, being cost-effective and more efficient, it provides a safer triaging approach, particularly beneficial in resource-limited settings.
Wasted tea of Salvia officinalis (WTSO) and the acorn cupule of Quercus coccifera (ACQC) were used as highly efficient biosorbents in the experiments. The aim is to convert waste materials into novel treatment materials which will be economically cheaper sources compared with conventional activated carbon. The powdered form of these materials, without any thermal or chemical pretreatment, was applied to wastewater containing Blue X GRL (BXGRL) and Red Violet 3R (RV3R) textile dyes. Various parameters affecting the separation such as pH, amount of biosorbent, contact time, dye concentration, and temperature were investigated. High removal efficiencies (up to 99%) and biosorption capacities could be readily obtained by these novel biosorbents for the studied toxic dyes. The most convenient pH and adsorbent dosage were found to be 8 and 0.1 g/100 mL, respectively. The values obtained from WTSO material with RV3R dye were fitted to Freundlich isotherm model, while the results from other material-dye sets were fitted to Langmuir model at 25 °C. The data from both ACQC and WTSO biosorbents were very well conformed to pseudo-second order reaction kinetic model with a R2 value of 99% when compared with pseudo first order and Elovich models. The rate limiting step was found to be chemisorption according to the results of pseudo second order and intraparticle diffusion models. In adsorption studies using high concentrations, it was observed that the dye removal efficiency increased with temperature. The highest biosorption capacities for BXGRL were obtained as 344.83 mg/g with WTSO and 222.22 mg/g for ACQC at 25 °C according to the Langmuir isotherm model. The biosorption capacities of WTSO and ACQC were 147.06 and 106.38 mg/g, respectively, for RV3R dye at 25 °C as a result of Langmuir model.
To validate a simple, low-cost protocol for transporting viable Enterobacterales at ambient temperature using sterile filter paper, addressing logistical and financial constraints in resource-limited laboratories. Four Enterobacterales strains were inoculated onto sterile Whatman Grade 3 filter paper and stored at room temperature. Viability was assessed up to 185 days using culture and MALDI-ToF MS. The method was further validated through two international shipments of clinical isolates from Madagascar to France and La Réunion at ambient temperature. Recovered isolates were evaluated for viability, species identification, and compatibility with genomic sequencing. All reference strains remained viable for at least 53 days, with the Enterobacter cloacae complex surviving beyond 185 days. MALDI-ToF MS identification remained stable throughout storage. Recovery rates from international shipments were 97.4% and 89.4%. All recovered isolates from one shipment yielded high-quality Oxford Nanopore genome assemblies. Filter paper transport was substantially cheaper ($0.27 per sample) than conventional transport methods. Sterile filter paper is a robust, cost-effective solution for ambient-temperature transport of viable Enterobacterales, preserving samples for extended periods and enabling phenotypic and genomic analyses. This approach supports decentralized diagnostics and surveillance without reliance on cold-chain infrastructure.
Data collection in randomized trials is expensive and labor intensive. With the rise in ongoing pragmatic trials, the use of electronic medical records (EMR) as a source of data has increased. Although potentially faster and cheaper, EMR use can lead to errors. Therefore, to ensure accurate data collection and to avoid systematic errors we performed a study comparing automated data extraction (ADE) with manual data extraction (MDE). We performed a retrospective cohort study to compare the accuracy of ADE using Structured Query Language with MDE by blinded physicians from our EMR. We tested the interrater agreement and intraclass correlation coefficient of clinical baseline data and outcomes of a random sample of 30 patients admitted to the ICU, on mechanical ventilation, requiring opioids for analgosedation for an upcoming pragmatic clinical trial. Key data compared included, but not limited to, patient's demographics, laboratory and vital signs, daily morphine milligram equivalent (MME), days alive and free of mechanical ventilation, days alive and free of hospitalization, days alive and free of ICU, days alive and free of vasopressors, and death. Among 238 patients screened over 1-month period, 72 fulfilled inclusion criteria and 30 were randomly selected to be included in the evaluation. We blindly collected 1320 baseline data, 2160 categorical outcomes and 705 continuous outcomes for a total of 4185 data points. The intraclass correlation coefficient and the Cohen's Kappa were perfect or almost perfect for all data, including outcomes such as daily MME, days alive and free of mechanical ventilation, days alive and free of ICU and days alive and free of hospital with p < 0.001. Among all rechecked data, the ADE was correct in 53 (77.9%) of cases, while MDE in 15 (22.1%). The inaccurate data collected by ADE accounted for 0.36% of the total data-points. The performance of ADE had almost perfect agreement for all outcomes and when rechecking for disagreements, it was more accurate than MDE.
This study leverages near-infrared (NIR) spectroscopy to develop a green analytical approach for the rapid screening of birch sap adulteration. NIR spectral data of samples were acquired using an NIR spectrometer, followed by sequential preprocessing via standard normal variate (SNV) transformation, multiplicative scatter correction (MSC), first derivative (1st Der), and Savitzky-Golay (SG) smoothing filtering. The impact of distinct preprocessing strategies on model performance was systematically evaluated. Qualitative analysis results demonstrated that the dung beetle optimization algorithm-enhanced support vector machine (DBO-SVM) effectively discriminated between pure birch sap and samples with varying adulteration levels, achieving a discrimination accuracy of 98% after SNV preprocessing. In terms of quantitative analysis, the proposed hybrid model (CNN-transformer-ECA) integrating convolutional neural network (CNN), transformer, and efficient channel attention (ECA) mechanism outperforms traditional machine learning models and other deep learning architectures in predicting the concentration of adulteration. For water dilution adulteration, the R2 p and RMSEp of the model's prediction set were 0.9482 and 2.6382, respectively. For sucrose solution adulteration, the corresponding R2 p and RMSEP values were 0.9505 and 0.9086, respectively. A high RPD further confirmed the model's excellent fitting capability and generalization performance. Under the constraint of maintaining high quantitative accuracy, network slimming (NS) was employed to significantly enhance model efficiency and reduce model size, and the pruned lightweight model was deployed on an embedded development platform. This study marks the first application of NIR spectroscopy in birch sap quality inspection, providing a theoretical foundation and technical framework for the rapid, non-destructive quality monitoring of related liquid beverages. PRACTICAL APPLICATIONS: This research enables food companies to quickly test birch water purity using a simple light scan, ensuring product quality before it reaches stores. Consumers would benefit from more reliable natural beverages, as this technology helps prevent adulterated products from being sold. The method provides a faster, cheaper alternative to traditional lab testing for quality control.
While the acute phase of the COVID-19 pandemic has passed, understanding the economic barriers to diagnostic access remains critical for future pandemic preparedness and universal health coverage. Implementing efficient testing modalities is crucial to achieving optimal value for both clients and healthcare providers. This study examines the cost and affordability of various SARS-CoV-2 antigen rapid-diagnostic-test modalities in Nigeria, Malawi, and Zimbabwe from a client perspective, providing a blueprint for future diagnostic strategies in Sub-Saharan Africa. Testing was offered for free through professional testing and self-testing in government or NGO-led primary healthcare centers across all countries, and in community pharmacies and drug stores in Nigeria. Data were collected from October 2022 to May 2023 through a survey of a random sample of adults visiting participating sites. The survey collected patient costs, including transportation, medical and non-medical expenses, and productivity loss. Affordability was assessed by the incidence of catastrophic health expenditure (defined as costs exceeding 10% of household income). The unit patient cost of testing in Nigeria, Malawi and Zimbabwe was $4.2, $2.7 and $2.7, respectively. In Nigeria, testing in community pharmacies and drug stores was cheaper than in primary healthcare centers. Self-testing cost less than professional testing in Nigeria ($1.3 versus $9.8), but more in Zimbabwe ($3.2 versus $2.3). In Malawi, Nigeria and Zimbabwe 40.6%, 28.6%, and 5.7% of clients, respectively, faced catastrophic health expenditures. SARS-CoV-2 antigen testing imposes a significant financial burden on clients. Even "free" testing carries high indirect costs that threaten diagnostic equity. Diversified testing modalities, such as community pharmacies and drug stores, may offer lower-cost options for sustainable diagnostic integration.
This chapter describes how to utilize spin-spin interactions measured by electron paramagnetic resonance (EPR) to gain structural details of lipid-protein interactions in samples containing more than one type of paramagnetic molecules. Common in these methods is that certain types of continuous wave (CW) EPR signals (also called spectral displays) are recorded under partial microwave saturation, extending into the region of a non-linear dependence of some spectral features on the microwave power (hence, the name non-linear CW EPR), and spectral parameters are derived that are proportional to the longitudinal (T1) relaxation time, which is in turn highly sensitive to spin-spin interaction between the paramagnetic molecules. Although powerful pulsed EPR techniques exist to measure T1 (and also for the transversal relaxation time, T2) directly, the advantage of the methods described here is that they can be performed on the much cheaper and more widespread conventional CW EPR instruments. During the 1990s, we developed and refined T1-sensitive nonlinear CW EPR techniques, which were then combined with paramagnetic quenching agents and doubly spin-labeled samples for distance measurements. In the following sections, first the concept of non-linear CW EPR will be introduced, then the common materials and methods will be described, followed by four example protocols and data analysis, and the chapter is concluded with useful notes. This chapter is incremental to the previous one in the sense that several experimental details are the same, such as the sources and types of spin-labeled lipid analogues, preparation of lipid-labeled bio- and model membranes, assaying lipid and protein content, sample geometry, basic instrument settings and processing of the EPR spectra (e.g., removal of the background and disturbing spectral components, normalization). Here, the focus is on the extra experimental and theoretical procedures required for measuring various types of spin-spin interactions and utilizing them for proximity relations in membranes. The example procedures and figures are illustrations, rather than exact reproductions of data from our previous works.
Diet affordability is a critical determinant of food security, health and wellbeing. However, the cost and affordability of diets have not been routinely measured in Queensland (Australia) in over a decade. This study assessed the cost and affordability of healthy (based on national healthy eating guidelines) and habitual (less healthy, based on national reported intake) diets across six Queensland regions. Data were collected in 35 communities, over two years (2023 and 2024), using the evidence-based Healthy Diets Australian Standardised Affordability and Pricing protocol. Data were analyzed relative to a six-person intergenerational Aboriginal and Torres Strait Islander reference household. Results indicate that, across Queensland, healthy diet costs are above the threshold for food stress in Aboriginal and Torres Strait Islander households. On average, healthy diets were 30% cheaper than the habitual diet (which include alcohol and takeaway foods) but cost at least 26% of household income (above the 25% threshold for food stress). In 2023, healthy diets were on average 31% more expensive in remote communities compared to urban and regional centers. In 2024, the cost of a healthy diet in remote communities decreased significantly by 24%, narrowing diet cost differences between remote and non-remote regions. This shift could be associated with the implementation of a freight subsidy in remote Queensland, or other influences on remote food pricing. Findings highlight diet-related cost-of-living challenges for Aboriginal and Torres Strait Islander families, underscore the need for ongoing monitoring and provide insight for policy interventions (such as targeted subsidies) to improve diet affordability and reduce nutrition-related health inequity.
Alzheimer's disease is the most common neurodegenerative disorder. Its pathological development is connected with the misfolding and accumulation of two toxic proteins: amyloid-beta and tau proteins. Mathematical models provide a valuable quantitative tool for monitoring disease progression. In this work, we proposed and compare a novel framework where the spatio-temporal dynamics of amyloid-beta and tau proteins is modeled based on employing either three-dimensional patient-specific geometries or through reduced network-based models defined on the brain connectome. More specifically, a high-fidelity biophysical model is proposed on three-dimensional brain geometries reconstructed from magnetic resonance imaging, whereas a network-based reduced formulation is defined on the brain connectome. For both approaches, a suitable numerical discretisation is proposed. A sensitivity analysis is presented to quantify the influence of model parameters on protein concentration patterns as well as compare the quality of the predictions. For both approaches, the results are validated against PET-SUVR clinical data using 18FAZD4694 for amyloid-beta and 18FMK6240 for tau protein. The results indicate that the three-dimensional model provides the most accurate and biologically consistent description of the disease progression, but remains computationally demanding. On the other hand, the reduced graph-based model is cheaper, but it is not always able to achieve reliable results.
Esketamine nasal spray (EN) has been approved in Germany since 2021 for the treatment of treatment-resistant depression (TRD). Historically, the development of EN resulted from the positive results of randomized clinical trials on the off-label use of subnarcotic ketamine infusions (SKI) for TRD. How effective, tolerable, and safe is off-label SKI compared to EN? Narrativ review. Selective literature search in PubMed for clinical studies comparing SKI and EN for TRD. International guidelines for depression and the cost-effectiveness of SKI and EN in inpatient treatment were also considered. Randomized direct comparative studies between EN and SKI are currently lacking. There is solid circumstantial evidence for the equivalence of acute treatment of TRD with SKI (usually 0.5 mg/kg body weight intravenously over 40-45 minutes) or EN (usually 56 or 84 mg per dose) in terms of effectiveness, safety, and tolerability: 9 meta-analyses (2 network meta-analyses) and 7 real-world studies (1 prospective, 6 retrospective comparative studies). SKI and subanesthetic esketamine infusions were also equivalent. A similar picture is emerging for serial treatment. Favorable tolerability and safety data have been available for EN for over 5 years (for serial SKI no such long-term observations). The USA and Canada have included off-label SKI as an alternative to approved EN in their guidelines for add-on treatment of TRD. For the German healthcare system, inpatient add-on treatment with SKI is about 2.3 to 3.8 times cheaper than with EN. Based on the above evidence, we estimate that there is currently low to moderate confidence according to GRADE for assuming the equivalence in effectiveness, tolerability, and safety of add-on treatment of TRD with SKI or EN. A more accurate assessment will be possible in 2030 at the earliest, when the results of an ongoing direct comparative clinical study are available. When updating the guidelines, we recommend checking whether the off-label add-on SKI can be provisionally placed at the same level as the add-on EN in the TRD treatment algorithm, as is the case in the USA and Canada. We hereby also encourage (i) health insurance companies to cover the costs of add-on SKI and (ii) the German authorities to review the off-label prescribability of SKI for TRD by the Joint Federal Committee (GB-A). Esketamin Nasenspray (EN) ist in Deutschland seit 2021 zur Behandlung der therapieresistenten Depression (TRD) zugelassen. Historisch resultierte die Entwicklung von EN aus den positiven Ergebnissen randomisierter klinischer Studien zum off-label Gebrauch von subnarkotischen Ketamin-Infusionen (SKI) gegen TRD. Wie wirksam, verträglich und sicher ist off-label SKI im Vergleich zu EN?Narratives Review. Selektive Literatursuche in PubMed bezüglich klinischer Studien, die sich mit dem Vergleich von SKI und EN gegen TRD befasst haben. Auch internationale Depressions-Leitlinien und die Wirtschaftlichkeit von SKI und EN bei stationärer Behandlung wurden betrachtet.Randomisierte direkte Vergleichsstudien zwischen EN und SKI fehlen bisher. Es gibt solide Indizien für die Gleichwertigkeit der Akutbehandlung der TRD mit SKI (üblicherweise 0,5 mg/kg Körpergewicht über 40–45 Minuten intravenös) oder EN (üblicherweise 56 oder 84 mg pro Gabe) bezüglich Effektivität, Sicherheit und Verträglichkeit: 9 Metaanalysen (2 Netzwerk-Metaanalysen) und 7 Real-World-Studien (1 prospektive, 6 retrospektive Vergleichsstudien). Auch SKI und subnarkotische Esketamin-Infusionen waren gleichwertig. Ein ähnliches Bild zeichnet sich für die serielle Behandlung ab. Günstige Verträglichkeits- und Sicherheitsdaten existieren für EN inzwischen über 5 Jahre (bei seriellen SKI noch keine so langen Verlaufsbeobachtungen). Die USA und Kanada haben off-label SKI als Alternative zum zugelassenen EN in ihre Leitlinien zur add-on Behandlung von TRD aufgenommen. Für das deutsche Gesundheitssystem ist die stationäre Behandlung mit SKI etwa 2,3 bis 3,8mal billiger als die mit EN.Aufgrund der oben genannten Indizien schätzen wir, dass für die Annahme der Gleichwertigkeit der add-on Behandlung der TRD mit SKI oder EN momentan eine niedrige bis moderate Vertrauenswürdigkeit nach GRADE vorliegt. Eine exaktere Einschätzung ist frühestens 2030 möglich, wenn das Ergebnis einer laufenden direkt vergleichenden klinischen Studie vorliegt. Wir empfehlen bei der Leitlinien-Aktualisierung zu prüfen, ob im TRD-Behandlungsalgorithmus die off-label add-on SKI vorläufig auf die gleiche Stufe wie die add-on EN gestellt werden kann (wie schon in den USA und Kanada). Hiermit regen wir außerdem (i) die Befürwortung der Kostenübernahme von add-on SKI durch die Krankenkassen und (ii) die behördliche Prüfung (Deutschland) der off-label Verordnungsfähigkeit von SKI bei TRD durch den Gemeinsamen Bundesausschuss (GB-A) an.
Operative notes in electronic health records contain critical information for understanding surgical care, yet manual coding is time-consuming, costly, and inconsistent. Large language models (LLMs) promise to transform this process by automatically extracting detailed procedure information - a capability with significant implications for scaling clinical registries and advancing surgical research. Here, we conducted a large-scale evaluation of state-of-the-art LLMs for zero-shot structured information extraction from orthopedic clinical notes. Fourteen open-source and proprietary models were tested on 800 real operative notes, annotated by both an orthopedic surgeon and an administrator using a curated list of 74 procedure classes. We compared model outputs to human annotations, assessing accuracy and exploring the effects of model scale, reasoning capabilities, and prompt design. We find that across models, LLMs consistently outperform administrator-assigned labels, achieving macro-F1 scores above 0.6 and improving over administrative coding by up to 10 points. Larger models and reasoning capabilities further boosted performance, though gains plateaued beyond 30 billion parameters. Performance varied by procedure frequency, revealing clear strengths and persistent challenges for rare or complex cases. Modern LLMs can already outperform routine administrative coding in extracting detailed surgical procedure data, pointing to a future where registry curation could be faster, cheaper, and more consistent. Yet, full alignment with surgical experts remains an open challenge- especially for rare procedures - emphasizing the need for domain adaptation and thoughtful deployment. Our findings illustrate how general-purpose LLMs can advance automated clinical data curation and inform the next generation of surgical informatics.
Traditional fluorescence polarization immunoassay (FPIA) requires costly monoclonal antibodies. This work develops a cost-effective fluorescence polarization enzyme assay (FPEA) for quercetin, employing α-amylase as a significantly cheaper recognition agent. Based on competitive binding with a fluorescent zearalenone tracer, the assay achieves a detection limit of 1.7 mg/mL, a working range of 2.3-6.4 mg/mL, and completes analysis within 5 min, offering a substantial time saving versus typical HPLC runs (~ 20 min). It demonstrates high specificity, with only minimal cross-reactivity to rutin (0.1%) and none to dihydroquercetin, gallic acid, and acarbose. FPEA analysis yielded a quercetin*α-amylase binding affinity constant of 3.9 × 103 L/mol. Successful application to onion peel extract and a dietary supplement yielded results consistent with liquid chromatography with diode-array detection (LC-DAD). The proposed FPEA provides a rapid, simple, and economical alternative for quality control of sources with high quercetin content.