搜索结果：original

This is a summary of the original research article "Aflibercept 8 mg versus Faricimab Treat‑and‑Extend for Diabetic Macular Edema or Neovascular Age‑Related Macular Degeneration: A Bayesian Fixed‑Effect Network Meta‑analysis of Clinical Trials." Diabetic macular edema (DME) and neovascular age-related macular degeneration (nAMD) are eye diseases that can be treated with anti-vascular endothelial growth factor (anti-VEGF) therapies. As these therapies are given by injections into the eye, reducing the number of injections could reduce the treatment burden on patients and healthcare providers. Aflibercept 8 mg and faricimab treat-and-extend (T&E) may enable longer dosing intervals compared with other available anti-VEGF therapies for patients with DME and nAMD; however, they have not been compared against each other directly in clinical trials. This study used a statistical method called a network meta-analysis (NMA) to compare aflibercept 8 mg and faricimab T&E by synthesizing data from clinical trials, enabling an indirect comparison in the absence of head-to-head studies. Outcomes included the mean number of injections, changes in vision [best-corrected visual acuity (BCVA)] and changes in macula thickness [central subfield thickness (CST)]. CST should reduce if the treatment is working. On the basis of indirect comparison of available clinical trial data, this exploratory study found that there were significantly fewer injections with aflibercept 8 mg treatment compared with faricimab T&E over 2 years, for both patients with DME and nAMD, and there were no significant differences between the two medications for BCVA or CST changes. NMAs are based on indirect comparisons and therefore have limitations, and clinical trials or real-world studies would be required to confirm these conclusions. Please refer to the original article for further details on the methods and limitations of this analysis.

Natural History Collections (NHC) often contain exceptional specimens documenting unusual historical events, and these records can be valuable for detecting overlooked biological introductions. Here, we examine a striking case: a Neotropical leaf beetle, Acentroptera norrisii, labelled as collected in central Spain in 1998 by Dr J. Modolell. Given the species' native distribution in the Neotropics, this record raises the question of whether it represents a failed historical introduction or a curatorial artefact. We reconstructed the circumstances of the collection event using the collector's personal archives. We conducted scanning electron microscopy (SEM) to document microscopic particles on the specimen's cuticle. We also analysed the content of the original specimen box to evaluate the likelihood of pollen cross-contamination, considering the geographical origin and family-level identity of all co-stored specimens. Archival evidence confirms the collector's presence at the labelled locality and date. SEM revealed several pollen grains consistent with plant taxa occurring at the collection site. However, the original box also contained numerous Spanish specimens belonging to beetle families with known anthophilous habits, meaning that cross-contamination during storage cannot be fully excluded. As a result, the palynological evidence remains inconclusive. Together, the available evidence makes a failed historical introduction a plausible scenario, yet the inconclusive particle analysis prevents any confident confirmation. This case highlights both the potential and the limitations of NHC-derived data in invasion biology: while museum collections can preserve traces of otherwise "invisible" introductions, interpreting isolated and context-poor specimens remains inherently uncertain.

Accurate modelling of tropical cyclones (TCs) is essential for reliable storm surge simulations, as TCs provide the primary external forcing. While deep learning models have demonstrated effectiveness in TC modelling, a challenge remains regarding the limited diversity of TCs in reanalysis data used for training-specifically, the scarcity of extreme TC events. This study proposes a physics-based data augmentation method that utilizes a numerical weather prediction (NWP) model to physically generate TCs lying beyond the range of reanalysis data. Subsequently, focusing on small-sized yet intense TCs as extreme cases, a model-originally pre-trained solely on reanalysis data-was fine-tuned using this augmented dataset to convert a parametric TC model (PM) field into an NWP-like field. Validation using test data mimicking extreme TCs and a storm surge hindcast of TC Faxai (a compact, intense TC that struck Tokyo Bay in 2019) revealed that the PM failed to simulate storm surges where topographic effects on the wind field are significant, and the pre-trained model underestimated wind speeds and storm surges. In contrast, the fine-tuned model successfully captured the spatiotemporal features of the extreme TCs and the peak storm surges upon TC landfall, achieving the lowest RMSE for storm surges across all TCs and tide gauges. These results suggest that physics-based data augmentation can effectively extend the applicability of deep learning models for TC and storm surge modelling to extreme events.

ObjectivesThe aims of the study were to describe a technique for feline ala vestibuloplasty and report long-term postoperative outcomes in 27 brachycephalic (BC) cats.MethodsThe study was a retrospective, questionnaire-based study of owned, clinically affected BC cats (n = 27). Owners completed a single, two-part questionnaire regarding presence or absence, frequency or severity of clinical signs before and >6 months after ala vestibuloplasty. Questions regarding respiratory and gastrointestinal signs, sleep and activity-related behaviors, and ocular, aural and dental health before and after surgery comprised the questionnaire. Response options were scored and individual scores summed to give a total clinical severity score for each cat (range; 0-129). Relationships between clinical variables and outcome were evaluated by regression analysis. Thirteen cats in the cohort were also enrolled in a previous questionnaire-based study reporting real-time preoperative clinical signs. Owner responses regarding preoperative clinical signs gathered at that time and in the present study were compared to assess for recall bias.ResultsTwenty-five/27 (92.5%) cats sustained long-term clinical improvement a mean of 1190 days (range; 354-1802 days) following ala vestibuloplasty. Median clinical severity score improved from 48 preoperative to 17 postoperative (p<0.0001). Owners reported reduced frequencies of sneezing (p<0.0001), coughing (p=0.001), nasal discharge (p=0.0002), snoring (p<0.0001), open-mouth breathing (p=0.001), dyspnea (p=0.0001), difficulty eating (p=0.04), messy eating (p=0.01), dyspnea while eating (p=0.003), regurgitation (p=0.03), vomiting (p=0.03) and dyspnea during activity (p=0.007). Owners also reported increased frequency (p=0.0002) and duration (p=0.0004) of activity. Owners of cats enrolled in both studies recalled their cats' preoperative clinical signs as slightly more severe than originally reported.Conclusions and relevanceAla vestibuloplasty yields long-term clinical improvements in cats' respiratory, gastrointestinal, sleep, activity, aural and ocular signs. Ala vestibuloplasty should be considered for effective treatment of the described clinical signs of feline brachycephalic syndrome.

Diabetes during pregnancy, whether gestational diabetes mellitus (GDM) or pre-existing diabetes mellitus (DM), is associated with increased maternal and neonatal risks. Remote care has shown promise in managing diabetes in the general population, however, its impact on the quality of antenatal care for pregnant women with diabetes remains unclear. Considering the growing adoption of remote care globally, it is vital to understand its impact on the quality of care. This review aims to evaluate the impact of remote care on the quality of care for pregnant women with diabetes. This manuscript describes the protocol for the review. MEDLINE, EMBASE, CINAHL, MIDIRS, Scopus, Cochrane, and Global Health databases will be searched to identify studies published from 2005 to 2025. Eligible studies will include original empirical research, encompassing both interventional (e.g. randomised controlled trials) and observational designs (e.g. cohort and cross-sectional studies), assessing remote care, including both virtual consultations and remote monitoring, for pregnant women with GDM, Type 1 or 2 DM. Quality of care will be assessed using the Institute of Medicine healthcare quality framework: patient-centredness, effectiveness, safety, efficiency, timeliness, and equity. Screening and data extraction will be conducted by two independent reviewers. The quality of the studies will be assessed using the Cochrane Risk of Bias tools. Subgroup analyses will be undertaken to explore variation by diabetes and intervention type. This protocol follows the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocol (PRISMA-P) guideline. The findings of this review will provide evidence on the impact of remote care on the quality of care for pregnant women with pre-existing DM and GDM. Strengthening the evidence will support the development of evidence-based implementation strategies and inform policy decisions to ensure safe, effective, and patient-centred use of remote care interventions. PROSPERO CRD420251024685.

Cortisol measurements are often performed as part of dynamic testing for endocrine disorders. However, considerable variation between cortisol assays may prevent the interchangeability of dynamic test results and interpretation. This study evaluated the comparability and inter-laboratory performance of cortisol immunoassays with that of analytically preferred LC-MS/MS cortisol assays. Serum samples from 45 participants were analyzed using Roche Cobas, Abbott Architect/Alinity, Siemens Atellica (including Siemens Atellica CSI), Beckman Coulter DxI/access immunoassays, and LC-MS/MS in 4-6 laboratories per assay. Passing-Bablok regression and Bland-Altman analyses were performed and inter-laboratory performance was determined for each immunoassay. Roche, Abbott and Beckman Coulter immunoassays showed good agreement with LC-MS/MS (bias&#xa0;-3.2&#x202f;,&#xa0;-6.7&#x202f; and&#xa0;-5.6&#x202f;%). The original Siemens immunoassay showed poor agreement (bias 25.3&#x202f;%), which improved substantially with the Siemens CSI assay (bias 4.9&#x202f;%). Inter-laboratory performance was acceptable for Roche and Abbott, intermediate for Siemens CSI, but poor for the Beckman Coulter assay. Respectively 0, 2.3, 12.5 and 56.8&#x202f;% of the cortisol measurements exceeded the maximal allowable imprecision. Assay comparability and inter-laboratory performance were acceptable for the Roche and Abbott cortisol immunoassays. The Beckman Coulter immunoassay showed high imprecision necessitating improvement. The bias of the Siemens immunoassay vs. LC-MS/MS was strongly reduced following immunoassay improvement (Siemens CSI), demonstrating that significant efforts can lead to much-needed assay improvements. Based on our findings using human sera primarily within the basal cortisol concentration range, we conclude that there is no need for a national standardization program for serum cortisol measurements.

Wild-type epidermal growth factor receptor (EGFRwt) is commonly implicated in tumor growth, yet most of the approved EGFR-targeted therapies are against mutant receptor isoforms, and patients with EGFRwt-dependent cancers have limited treatment options. We herein explored the potential to inhibit EGFRwt with five top screened hits retrieved from Food and Drug Administration (FDA)-approved kinase inhibitors library, originally developed for MET-overexpressing cancers for potential drug repurposing approach. Induced-fit docking showed that all five molecules bound the ATP-binding pocket of EGFRwt, with compound D4 emerging as the top candidate due to key interactions with ASP855 (DFG motif) and MET793 (hinge region), underscoring its favorable engagement. To further assess this interaction, we then conducted 100 ns molecular dynamics (MD) simulations, confirming the structural stability of the EGFRwt-D4 complex, with stable root-mean-square deviation (RMSD) values and maintained secondary structure elements. Free energy of binding calculations using the Molecular Mechanics/Poisson-Boltzmann Surface Area (MM/PBSA) method supported these findings, with D4 showing the most favourable interaction, dominated by Van Der Waals and electrostatic contributions from key catalytic amino acid. Our research, with further experimental validation, could be helpful to support that certain MET inhibitors, especially D4, are promising competitive inhibitors of EGFRwt, offering potential for the development of new therapeutic strategies targeting EGFRwt-driven cancers.

Arrowroot starch is resistant to amylase, hindering its porous modification via conventional enzymatic approaches. Herein, two eco-friendly non-enzymatic methods, gel cooling-solvent exchange and gel freezing-solvent exchange, were developed to prepare porous arrowroot starch. Retrogradation time at 4&#xa0;&#xb0;C was systematically evaluated. Starch retrograded for 72&#xa0;h presented uniform interconnected mesopores with a specific surface area of 71.61&#xa0;m2/g, superior to the gel-frozen counterpart. Spectroscopic results revealed no new functional groups, and modified starch maintained original thermal stability. Its water and oil adsorption capacities were significantly enhanced. Pore formation relied on the synergistic effect of moderate retrogradation and solvent exchange. This study provides a practical strategy for porous modification of amylase-resistant starches, promising broad applications in food and environmental fields.

Lymph node metastasis (LNM) is a critical clinical indicator for determining the initial treatment strategy for patients with lung cancer. However, accurately diagnosing LNM preoperatively remains a significant challenge. Data-driven predictive modeling has become a mainstream approach to address this issue, yet it often overlooks existing clinical knowledge. Large language models (LLMs) have demonstrated the potential to predict clinical risks in a zero-shot manner based on the extensive clinical knowledge learned from large-scale corpora. LLMs have demonstrated the potential to predict clinical risks in a zero-shot manner based on the extensive clinical knowledge learned from large-scale corpora. This study aims to investigate the integration of LLM-derived knowledge with data-driven patterns to enhance the accuracy of LNM prediction. We propose a novel ensemble framework that combines the strengths of LLMs and machine learning (ML) models for LNM prediction in lung cancer. Specifically, 3 ML models were trained using clinical data, and their predicted probabilities, along with the original clinical features, were incorporated into prompts for LLMs. Three LLMs-GPT-5.4, GPT-5.4-nano, and DeepSeek-V3.2-were used to independently predict LNM risk 5 times, and 4 ensemble strategies were applied to aggregate their predictions into a final outcome. The proposed approach was evaluated on clinical data from 767 patients with lung cancer at Peking University Cancer Hospital. Experimental results show that our proposed framework significantly outperforms base ML models, achieving an area under the curve of 0.781 and an average precision of 0.420. Compared with the no reasoning English setting, both the reasoning English setting and nonreasoning Chinese setting showed a lower area under the curve but higher average precision. This study presents a novel knowledge-augmented strategy for integrating the clinical knowledge embedded in LLMs with the statistical patterns captured by ML models to improve the LNM prediction of lung cancer, offering a new paradigm for integrating medical knowledge and patient data in clinical predictions.

Early diagnosis is an effective strategy in chronic obstructive pulmonary disease (COPD) prevention. Active case-finding is an effective approach, but traditional tools such as COPD-SQ are limited by outdated data, poor extrapolation, and singular binary prediction. This study aimed to develop an updated, convenient, and interpretable machine learning tool for COPD screening in community participants. Data for model training and external validation were obtained from two community-based studies in Guangdong, China. PyCaret and R programming language were used to develop machine learning models. Thirty original items, including demographic data, clinical features, and risk factor data, were initially used. Eleven machine learning classification models were compared, and the least absolute shrinkage and selection operator was further used to shrink predictors. Model performance was evaluated using ROC, AUC, accuracy, sensitivity, specificity, and other metrics. Shapley Additive exPlanations were used to interpret the models. A total of 5381 and 2456 participants from the training and external validation cohorts were included, respectively. In predicting COPD, the AdaBoost model showed the best performance, with an accuracy of 0.846 and an AUC of 0.848. For GOLD classification prediction, the model achieved an overall accuracy of 0.822 and an AUC of 0.816, and identified 83% of moderate-to-severe COPD in the community. In regression analysis, the gradient boosting regression model showed good consistency between predicted and measured FEV1 %pred and FEV1/FVC values. The models also demonstrated good performance in the external validation cohort and were deployed online. We constructed an active case-finding tool with integrated machine learning models for predicting COPD, COPD severity, and lung function parameters using limited clinical data. This tool may help prioritize high-risk individuals for confirmatory spirometry in community settings. Future implementation studies should evaluate its effect on referral efficiency, diagnostic yield, treatment uptake, and long-term outcomes.

Single-cell sequencing has revolutionized biomedical research by offering insights into cellular heterogeneity at unprecedented resolution. Yet, the low signal-to-noise ratio characteristic of single-cell RNA sequencing (scRNA-seq) challenges quantitative analyses. Gene regulatory network (GRN) analysis can help overcome this obstacle, enabling the mechanistic elucidation of cellular state determinants. For instance, the VIPER algorithm can identify Master Regulator proteins from gene expression data. However, as the size and complexity of scRNA-seq datasets grow, the demand for scalable tools supporting the analysis of datasets with up to hundreds of thousands of cells becomes increasingly critical in its original implementation in R. RESULTS: To address this challenge, we introduce pyVIPER, a Python-based tool for protein activity inference from transcriptional data. pyVIPER supports flexible data transformation/postprocessing modules, enrichment analysis algorithms, and features a novel data structure for GRNs manipulation. It integrates seamlessly with scverse, scanpy and widely adopted machine learning libraries. By leveraging PyTorch-based GPU acceleration and optimized core operations, benchmarking demonstrates orders-of-magnitude improvements in runtime efficiency compared to R-based VIPER, reducing analysis time for large datasets from hours to minutes. CONCLUSIONS: pyVIPER is a fast, memory-efficient, and highly scalable Python toolkit for protein activity inference in large-scale scRNA-seq datasets. Its scalability and hardware acceleration enables high-throughput VIPER-based analysis of virtually any single-cell dataset while facilitating integration with other Python-based, including state-of-the-art machine learning workflows. Taken together, these features make pyVIPER a valuable resource to expand the applicability of mechanistic regulatory network-based analysis in single-cell research.

The Restrictive versus Liberal Fluid Therapy in Major Abdominal Surgery (RELIEF) randomised trial found that a restrictive intravenous fluid regimen increased the risk of acute kidney injury (AKI) after major abdominal surgery. We examined whether AKI after restrictive fluid therapy increased the risk for chronic kidney disease (CKD). We conducted long-term follow-up between 90 days and 48 months after surgery in RELIEF participants randomly assigned to a restrictive vs liberal fluid regimen, from May 2013 through September 2016. The primary outcome was incident or progressive CKD (stage &#x2265;3), including worsening by &#x2265;1 CKD stage in participants with preoperative CKD stage &#x2265;3. Secondary outcomes included lowest recorded estimated glomerular filtration rate (eGFR) and maximal change in eGFR from baseline. Long-term follow-up data were obtained for 1670/2983 participants (mean age 69 (range: 58-75)yr; 47% females) included in the modified intention-to-treat analysis of the original RELIEF trial. The primary analysis was performed for 754 (49.6%) assigned to a restrictive fluid regimen and 766 (50.4%) assigned to a liberal fluid regimen. The incidence of new or worse CKD &#x2265; stage 3 did not differ between 422/754 (56%) participants assigned to restrictive fluid therapy, compared with 409/766 (53.4%) participants who received liberal fluid therapy (adjusted odds ratio 1.13 [95% CI 0.91-1.41]). These findings remained unaltered in sensitivity analyses. There were also no differences for eGFR secondary outcomes. Within the limits of this post hoc analysis, no difference was found in the incidence of new or progressive CKD was not related to a restrictive perioperative fluid regimen. ClinicalTrials.gov (NCT01424150).

Gait patterns in children with myelomeningocele (MMC) at various neurological levels have been described, both with and without orthotic support. Although the neurological level of the lesion serves as an important predictor of ambulatory potential, the expected walking ability is not always achieved, as additional factors such as spasticity may influence gait negatively. The aim of this study was to retrospectively compare gait patterns as assessed in childhood with those observed in adulthood. Of 59 individuals with MMC aged 18 years or older, 29 had undergone three-dimensional gait analysis in childhood (Ch-GA). These data were retrospectively analysed and compared with findings from a subsequent adult gait analysis (Ad-GA). The mean (standard deviation) age at the time of Ch-GA was 11.6 (4.1) years and at Ad-GA 25.9 (3.9) years. The median (range) interval between assessments was 15.0 years (5.1-17.2). Twenty-two participants maintained independent, non-assisted walking (Group A), 5 had transitioned from independent walking to using a walking aid (Group B), and 2 used a walking aid at both Ch-GA and Ad-GA (Group C), with individualized orthotic prescriptions provided at both time points. In Group A, two of eleven kinematic variables and six of eleven kinetic variables in the hip, knee, and ankle showed deterioration, and walking speed had decreased. Functional ambulation declined from 18 community ambulators and 4 household ambulators (Ha) in childhood to 8 and 14, respectively, in adulthood. In Group B, analysed with only Gait Deviation Index (GDI), values were unchanged, but all temporospatial gait parameters had deteriorated. Functional ambulation decreased from five individuals classified as Ha to two Ha and three non-functional ambulators. The two individuals in Group C, who used a walker at both assessments, largely maintained the same GDI values and temporospatial parameters as in childhood. Largely consistent with our original expectations, the findings indicate that gait patterns remain relatively stable from childhood to adulthood in individuals with MMC when supported by appropriate rehabilitation interventions, though some deterioration of gait and ambulation occurred. The results reflect gait-related changes that can be expected during growth in this population.

Increased disease awareness and advanced serological tests have improved myasthenia gravis (MG) diagnosis. However, the risk of misdiagnosis remains. The aim of this study was to evaluate the rate and causes of MG misdiagnosis in a tertiary MG clinic. We analyzed medical records from patients diagnosed with MG who were referred to our tertiary clinic for a second opinion between 2019 and 2024. Of the 144 patients originally included in the analysis, 15 were excluded due to insufficient data. MG diagnosis was not confirmed in 26/129 patients (20.2%). None of these 26 patients had clinical symptoms consistent with MG; 13 patients (50%) were seronegative; 12 had AChR antibodies (Abs) (detected by radioimmunoassay -RIA in 6 and by ELISA in the other 6 patients); a single patient had MuSK-Abs. Fixed and live cell-based assays were negative in all these 26 cases. Predictors of MG misdiagnosis included absence of MG-typical clinical features, younger age at onset, female sex, non-response to or intolerance of pyridostigmine, Ab testing by ELISA. MG misdiagnosis is not uncommon and often leads to unnecessary treatment. Our findings confirm the relevance of typical clinical pattern, proper antibody assays, and expertise in neurophysiological testing to establish MG diagnosis.

Cerebral venous thrombosis (CVT) predominantly affects young adults and frequently presents with headache. Although functional outcomes are generally favorable, many patients report persistent or recurrent headache long after the acute event. This systematic review aimed to assess the prevalence, clinical characteristics, associated factors, and management of long-term headache following CVT. A systematic search of PubMed and Embase (2000-2025) identified studies reporting original data on long-term headache after CVT in adult populations. Case series including fewer than 30 patients were excluded. Given substantial heterogeneity in study design, definitions, and outcome measures, a qualitative synthesis was performed. Fifteen studies including 2,136 patients were analyzed. The reported prevalence of post-CVT headache ranged from 14% to 59%. Headache most often resembled primary headache disorders, particularly tension-type and migraine-like phenotypes. Available data suggest that de novo headache may represent the most frequent presentation, although modification of pre-existing headache disorders also occurs. Pre-existing headache and depression were associated with post-CVT headache, whereas intracerebral hemorrhage was not. Associations with venous recanalization, thrombosis location, and anticoagulation delay were inconsistent. Secondary mechanisms, including intracranial hypertension, dural arteriovenous fistula, CVT recurrence, and medication-overuse headache, were identified in some patients. Data on management strategies were limited. Long-term headache is a frequent but under-recognized complication of CVT that may significantly impact quality of life. The absence of a standardized definition contributes to heterogeneity across studies. After exclusion of secondary causes, post-CVT headache encompasses both de novo and modified primary headache phenotypes. A consensus term such as "new or modified recurrent headache attributed to past CVT" may help describe this spectrum and improve clinical characterization and future research.

The uniformity of tobacco blend mixing is a critical factor influencing cigarette quality. However, due to the dynamic nature of the blending process and the complex characteristics of the materials involved, current detection methods primarily rely on manual sampling or offline analysis. These approaches suffer from poor real-time performance and lack representativeness. To address these limitations, this study explores methods for generating tobacco-leaf image data and proposes a real-time semantic segmentation technique for tobacco-leaf images based on U-Net. First, a small set of tobacco leaf images was acquired to establish a raw dataset of composite tobacco leaf images. Three generative adversarial networks-CycleGAN, WGAN, and DCGAN-were employed to generate tobacco leaf images from this dataset. Based on image quality, the optimal image generation network was selected. Experimental results showed that CycleGAN produced the highest-quality tobacco leaf images. Next, the Squeeze Excitation Attention (SE) mechanism was integrated into the VGG16 backbone network. This enhancement improved the network's ability to extract features from various types of tobacco leaf images while suppressing interference from irrelevant pixels. The results demonstrated that, compared to mainstream models such as Segformer, DeepLabV3, PSPNet, and the unmodified U-Net, the proposed SE-UNet model achieved superior segmentation accuracy, excellent real-time performance, and overall optimal segmentation capabilities. Compared to Segformer, DeepLabV3, PSPNet, and the original model, the MIOU in the evaluation metrics improved by 17.46, 8.7, 13.39, and 2.77 percentage points, respectively. Finally, the pixel areas of different tobacco leaf types were extracted and calculated from the segmented images to determine the proportion of each leaf type within the images. This research offers a novel approach for practical tobacco production and quality inspection, while also providing a new pathway for real-time online detection of other agricultural products.