Search Results (1,134)

Background: Coronary artery bypass grafting (CABG) is a well-established revascularization strategy for patients with multivessel coronary artery disease. The effectiveness of CABG is significantly influenced by antiplatelet therapy aimed at maintaining graft patency and reducing thrombotic complications. However, substantial inter-individual variability exists in platelet function responses to standard therapies such as aspirin and clopidogrel, leading to antiplatelet resistance. This variability has been linked to increased risks of myocardial infarction, stroke, and early graft failure. Platelet function testing (PFT) offers a potential strategy to identify resistance and guide more personalized antiplatelet therapy. This study aims to evaluate the association between perioperative platelet function test results and clinical outcomes following CABG. By assessing platelet responsiveness at multiple timepoints and correlating findings with postoperative events, the study seeks to determine whether PFT can stratify risk and improve patient management. Methods: This is a prospective, single-centre, observational cohort study conducted at a tertiary NHS cardiac surgery centre. Patients having elective or urgent isolated CABG will be enrolled and undergo perioperative PFT using the TEG6s system. Clinical outcomes will be monitored for 12 months postoperatively, with primary endpoints assessing the correlation between platelet function results and major adverse cardiovascular and cerebrovascular events (MACCE). Secondary endpoints will include the prevalence of antiplatelet resistance, demographic predictors, and the feasibility of integrating PFT into clinical workflows. Results: This study will report the prevalence of aspirin and clopidogrel resistance in CABG patients based on TEG6s PFT, as well as the correlation between platelet function results and MACCE, postoperative bleeding, and the need for surgical re-exploration. Additionally, it will examine the associations between demographic and clinical factors—such as diabetes status, renal function, BMI, and surgical technique—and variability in platelet responsiveness. The feasibility of incorporating PFT into perioperative workflows will also be evaluated, assessing whether results could support personalized antiplatelet management in future clinical trials. Conclusions: Findings from this study will provide real-world evidence regarding platelet function variability in CABG patients and suggest that PFT may identify those at increased risk of thrombotic complications. This exploratory analysis supports the need for larger interventional trials aimed at optimizing individualized postoperative antiplatelet therapy to improve surgical outcomes. Full article

Monitoring experimental animals is essential for ethical, scientific, and financial reasons. Conventional observation methods are limited by subjectivity and time constraints. Camera-based monitoring combined with machine learning offers a promising solution for automating the monitoring process. This study aimed to validate and assess the performance of a machine learning model for analyzing drinking behavior in minipigs. A novel, vision-based monitoring system was developed and tested to detect drinking behavior in minipigs. The system, based on low-cost Raspberry Pi units, enabled on-site video analysis. A dataset of 5297 images was used to train a YOLOv11n object detection model to identify key features such as pig heads and water faucets. Drinking events were defined by the spatial proximity of these features within video frames. The multi-class object detection model achieved an accuracy of above 97%. Manual validation using human-annotated ground truth on 72 h of video yielded an overall accuracy of 99.7%, with a precision of 99.7%, recall of 99.2%, and F1-score of 99.5%. Drinking patterns for three pigs were analyzed using 216 h of video. The results revealed a bimodal drinking pattern and substantial inter-pig variability. A limitation to the study was chosen methods missing distinguishment between multiple pigs and the absence of quantification of water intake. This study demonstrates the feasibility of a low-cost, computer vision-based system for monitoring drinking behavior in individually housed experimental pigs, supporting earlier detection of illness. Full article

Functional electrical stimulation (FES) is widely used to improve gait in individuals with neurological impairments; however, early responses in adults with congenital conditions, such as cerebral palsy, who are newly exposed to FES, remain poorly understood. This study investigated the orthotic and therapeutic effects of FES in short- and long-term users using standardized three-dimensional gait analysis. In this longitudinal study, short-term users (G1; n = 13; mean age 31.7 ± 18.1 years) were evaluated both without and with FES and followed over a 4–12-week insurance-covered trial period. Long-term users (G2; n = 11; mean age 32.2 ± 11.0 years), who had used FES for at least one year, were reassessed over a standardized 12-week interval. Linear mixed-effects models assessed the effects of FES and time, with subjects included as random effects to account for inter-individual variability. G1 showed significant therapeutic adaptations, including increased walking speed and step length and reduced step width, accompanied by decreased dorsiflexion during stance and swing, while no significant orthotic effects were observed. G2 demonstrated clear orthotic responses, such as increased dorsiflexion at heel strike and during swing and improved walking speed and step length, with minimal evidence of additional therapeutic adaptation. The initial reduction in dorsiflexion in G1 warrants further investigation. These findings suggest that evaluation timelines may need to be extended and that outcome measures beyond foot clearance should be considered, particularly given the heterogeneity and severity of congenital neurological conditions. Full article

Background/Objectives: Childhood metabolic dysregulation exerts a profound influence on the development of obesity and metabolic diseases. The human gut microbiota, with highly personalized characteristics, plays an important role in host metabolism. However, the dynamics of gut microbial features during this developmental phase are still unclear. This longitudinal observational study collected 204 fecal samples and 153 blood samples from 51 children (aged 8.90 ± 0.78 years) at four timepoints over 52 weeks, aiming to identify dynamic changes in individual gut microbiota and underlying mechanistic interactions that predict measures of pediatric metabolic health. Methods: Fecal samples were subjected to 16S rRNA gene amplicon sequencing and short-chain fatty acid quantification. Serum samples were analyzed for biochemical tests. Dietary intake, physical activity, clinical phenotypes, early-life factors, and fecal characteristics were further assessed. Results: In the results, the fecal microbiota dynamics exhibit inter-individual variation among children, allowing classification into high- and low-stability subgroups based on intra-individual β-diversity variability. Children with low-stability microbiota had adverse blood lipid profiles (p < 0.05). Compared to the high-stability group, the low-stability microbiota demonstrated significant association with low dietary fiber and highly variable amino acid consumption (|r| > 0.3, q < 0.05). Low-stability microbiota exhibited marked fluctuations in Phocaeicola vulgatus, which was strongly linked to both blood triglycerides and lipoprotein(a) levels, as well as dietary fiber and amino acid intake. Baseline depletion of P. vulgatus and Faecalibacterium duncaniae, combined with the children’s physiological status, lifestyle behaviors, and early-life factors, predicted microbial stability classification (AUROC = 0.93). Conclusions: These findings suggested that the variation in the gut microbiota dynamics could be considered as a possible complementary biomarker to understand the individualized responses within dietary interventions aimed at improving metabolic health in childhood. Further well-designed intervention study is needed to define these observational associations. Full article

Immunohistochemistry (IHC) is essential for diagnostic, prognostic, and predictive biomarker assessment in oncology, but manual interpretation is limited by subjectivity and inter-observer variability. Machine learning (ML), a computational subset of AI that allows algorithms to recognise patterns and learn from annotated datasets to make predictions or decisions, has led to advancements in digital pathology by supporting automated quantification of biomarker expression on whole-slide images (WSIs). This review evaluates the role of ML-assisted IHC scoring in the transition from validated biomarkers to the discovery of emerging prognostic and predictive IHC biomarkers for genitourinary (GU) tumours. Current applications include ML-based scoring of routinely used biomarkers such as ER/PR, HER2, mismatch repair (MMR) proteins, PD-L1, and Ki-67, demonstrating improved consistency and scalability. Emerging studies in GU cancers show that algorithms can quantify markers including androgen receptor (AR), PTEN, cytokeratins, Uroplakin II, Nectin-4 and immune checkpoint proteins, with early evidence indicating associations between ML-derived metrics and clinical outcomes. Important limitations remain, including limited availability of training datasets, variability in staining protocols, and regulatory challenges. Overall, ML-assisted IHC scoring is a reproducible and evolving approach that may support biomarker discovery and enhance precision GU oncology. Full article

The marine environment in the South Pacific Island Countries (SPICs) is sensitive and vulnerable to climate change. While large-scale changes in this region are well-documented, national-scale analyses that address management needs remain limited. This study evaluated the performance of satellite-derived datasets—including sea surface temperature (SST), sea surface salinity (SSS), Secchi disk depth (SDD), chlorophyll-a (Chl-a), net primary production (NPP), and sea level anomaly (SLA)—against in situ observations, and analyzed their spatial and temporal variability across 12 national Exclusive Economic Zones (EEZs) during 1998–2023. Validation results presented that current satellite datasets could provide applicable information for EEZ-scale analyses. In the past decades, the SPICs experienced a general increase in SST and SLA, accompanied by marked within-EEZ heterogeneity in Chl-a and NPP variations, with Papua New Guinea exhibiting the largest within-EEZ inter-annual variability. In addition to monitoring, satellite data would help to constrain the uncertainty of CMIP6 results in the SPICs, subject to the accuracy of specific products. By 2100, Nauru might experience the most vulnerable EEZ, while the marine environment in the French Polynesian EEZ can keep relatively stable among all 12 EEZs. Meanwhile, CMIP6 projections in the Southeastern EEZs are more sensitive to satellite-based constraints, showing pronounced adjustments. Our results demonstrate the potential of combining validated satellite data with CMIP6 models to provide national-scale decision support for climate adaptation and marine resource management in the SPICs. Full article

Accurate multivariate traffic flow forecasting is critical for intelligent transportation systems yet remains challenging due to the complex interplay of temporal dynamics and spatial interactions. While Transformer-based models have shown promise in capturing long-range temporal dependencies, most existing approaches compress multidimensional observations into flattened sequences—thereby neglecting explicit modeling of cross-dimensional (i.e., spatial or inter-variable) relationships, which are essential for capturing traffic propagation, network-wide congestion, and node-specific behaviors. To address this limitation, we propose TSAformer, a novel Transformer architecture that explicitly preserves and jointly models time and dimension as dual structural axes. TSAformer begins with a multimodal input embedding layer that encodes raw traffic values alongside temporal context (time-of-day and day-of-week) and node-specific positional features, ensuring rich semantic representation. The core of TSAformer is the Two-Stage Attention (TSA) module, which first models intra-dimensional temporal evolution via time-axis self-attention then captures inter-dimensional spatial interactions through a lightweight routing mechanism—avoiding quadratic complexity while enabling all-to-all cross-node communication. Built upon TSA, a hierarchical encoder–decoder (HED) structure further enhances forecasting by modeling traffic patterns across multiple temporal scales, from fine-grained fluctuations to macroscopic trends, and fusing predictions via cross-scale attention. Extensive experiments on three real-world traffic datasets—including urban road networks and highway systems—demonstrate that TSAformer consistently outperforms state-of-the-art baselines across short-term and long-term forecasting horizons. Notably, it achieves top-ranked performance in 36 out of 58 critical evaluation scenarios, including peak-hour and event-driven congestion prediction. By explicitly modeling both temporal and dimensional dependencies without structural compromise, TSAformer provides a scalable, interpretable, and high-performance solution for spatiotemporal traffic forecasting. Full article

Breslow thickness (BT) is the most powerful prognostic indicator in cutaneous melanoma, yet histopathological measurement exhibits some limitations such as interobserver variability and diagnostic delays. Preoperative clinical assessment demonstrates 30% misclassification rates. This narrative review synthesizes evidence on deep learning models for non-invasive BT estimation from dermoscopic images. Convolutional neural networks (ResNet, EfficientNet, Vision Transformers) with transfer learning from ImageNet achieve up to 75–79% accuracy and AUC 0.76–0.85 on single-center datasets. Preprocessing techniques (hair removal, color normalization, data augmentation) and interpretability methods (Grad-CAM, LIME) enhance clinical applicability. However, external validation reveals performance degradation. The clinically critical thickness range (0.4–1.0 mm) demonstrates poor discrimination. Significant dataset bias exists: most training data represents lighter skin phototypes, resulting in an underrepresentation of darker skin types. AI models function as complementary decision-support tools rather than replacements for histopathology. Prospective clinical trials validating clinical utility are lacking, and regulatory approval pathways are undefined. Research priorities include diverse public datasets with balanced skin tone representation, the adoption of threshold-weighted loss functions to prioritize accuracy at the 0.8 mm surgical cut-off, multi-institutional external validation, prospective randomized trials, federated learning frameworks, and regulatory engagement. Only rigorous, equitable research can translate AI from proof-of-concept to clinically reliable tools benefiting all melanoma patients. Full article

The DAIC-WOZ dataset is a widely used benchmark for the task of depression severity estimation from multimodal behavioral data. Yet the reliability, reproducibility, and methodological rigor of published machine learning models remain uncertain. In this systematic review, we examined all works published through September 2025 that mention the DAIC-WOZ dataset and report mean absolute error as an evaluation metric. Our search identified 536 papers, of which 414 remained after deduplication. Following title and abstract screening, 132 records were selected for full-text review. After applying eligibility criteria, 66 papers were included in the quality assessment stage. Of these, only five met minimal reproducibility standards (such as clear data partitioning, model description, and training protocol documentation) and were included in this review. We found that published models suffer from poor documentation and methodology, and, inter alia, identified subject leakage as a critical methodological flaw. To illustrate its impact, we conducted experiments on the DAIC-WOZ dataset, comparing the performance of the model trained with and without subject leakage. Our results indicate that leakage produces significant overestimation of the validation performance; however, our evidence is limited to the audio, text, and combined modalities of the DAIC-WOZ dataset. Without leakage, the model consistently performed worse than a simple mean predictor. Aside from poor methodological rigor, we found that the predictive accuracy of the included models is poor: reported MAEs on DAIC-WOZ are of the same magnitude as the dataset’s own PHQ-8 variability, and are comparable to or larger than the variability typically observed in general population samples. We conclude with specific recommendations aimed at improving the methodology, reproducibility, and documentation of manuscripts. Code for our experiments is publicly available. Full article

This contribution explores transgenerational language change in a historical migrant community by qualitatively examining the correspondence of first- and second-generation Scottish immigrants coming to New Zealand in the nineteenth century. Taking a microsocial approach, the letters of a migrant family and one other migrant are explored for language maintenance and shift, to identify whether Scots language features were lost altogether or continued to be utilised for specific social, personal and stylistic goals, despite the English-dominant space that the migrants operated in. In tandem, the adoption of early New Zealand English (NZE) and te reo Māori lexis is analysed, to identify differences in usage patterns that might point to different degrees of integration and mobility. Finally, inter-writer and inter-generational differences are examined in relation to the mobility and social networks of the correspondents, to consider how this might contribute to any variation observed. For the investigation, manuscript letters were digitised, and relevant features identified, extracted and discursively analysed. Results show the continuation of heritage features through a combination of style-oriented goals and learned letter-writing practices, while the adoption of new lexis is shown to occur within specific semantic domains that reflect the social mobility of the migrants. However, language maintenance and shift are not uniform between the writers, elucidating the highly variable experiences of migrants, even within the same family. Rather, contact-induced language changes are sensitive to minute differences across individuals, underpinning the value of nuanced explorations of historical migration and language change. Full article

Persistent cloud cover during the growing season and mosaic cropping patterns introduce temporal gaps and mixed pixels, undermining the reliability of large-scale crop identification and acreage statistics. To address these issues, we develop a high spatiotemporal-resolution remote-sensing approach tailored to heterogeneous farmlands. First, an improved Spatiotemporal Adaptive Reflectance Fusion Model (STARFM) is used to fuse Landsat, Sentinel-2, and MODIS observations, reconstructing a continuous Normalized Difference Vegetation Index (NDVI) time series at 30 m spatial and 8-day temporal resolution. Second, at the field scale, we derive phenological descriptors from the reconstructed series—key phenophase timing, amplitude, temporal trend, and growth rate—and use a Random Forest (RF) classifier for detailed crop discrimination. We further integrate SHapley Additive exPlanations (SHAP) to quantify each feature’s class-discriminative contribution and signed effect, thereby guiding feature-set optimization and threshold refinement. Finally, we generate a 2024 crop distribution map and conduct comparative evaluations. Relative to baselines without fusion or without phenological variables, the fused series mitigates single-sensor limitations under frequent cloud/rain and irregular acquisitions, enhances NDVI continuity and robustness, and reveals inter-crop temporal phase shifts that, when jointly exploited, reduce early-season confusion and improve identification accuracy. Independent validation yields an overall accuracy (OA) of 90.78% and a Cohen’s kappa(κ) coefficient of 0.882. Coupling dense NDVI reconstruction with phenology-aware constraints and SHAP-based interpretability demonstrably improves the accuracy and reliability of cropping-structure extraction in complex agricultural regions and provides a reusable pathway for regional-scale precision agricultural monitoring. Full article

Background: Recovery following incomplete spinal cord injury (iSCI) remains challenging, with conventional rehabilitation often emphasizing compensation over functional restoration. As most new spinal cord injury cases preserve some motor or sensory pathways, there is increasing interest in therapies that harness neuroplasticity. Robotic exoskeletons provide a promising means to deliver task-specific, repetitive gait training that may promote adaptive neural reorganization. This feasibility study investigates the feasibility, safety, and short-term effects of exoskeleton-assisted walking in individuals with degenerative iSCI. Methods: Two cooperative male patients (patients A and B) with degenerative iSCI (AIS C, neurological level L1) participated in a four-week intervention consisting of one hour of neuromotor physiotherapy followed by one hour of exoskeleton-assisted gait training, three times per week. Functional performance was assessed using the 10-Meter Walk Test, while gait quality was examined through spatiotemporal gait analysis. Vendor-generated surface electromyography (sEMG) plots were available only for qualitative description. Results: Patient A demonstrated a clinically meaningful increase in walking speed (+0.15 m/s). Spatiotemporal parameters showed mixed and non-uniform changes, including longer cycle, stance, and swing times, which reflect a slower stepping pattern rather than improved efficiency or coordination. Patient B showed a stable walking speed (+0.03 m/s) and persistent gait asymmetries. Qualitative sEMG plots are presented descriptively but cannot support interpretations of muscle recruitment patterns or neuromuscular changes. Conclusions: In this exploratory study, exoskeleton-assisted gait training was feasible and well tolerated when combined with conventional physiotherapy. However, observed changes were heterogeneous and do not allow causal or mechanistic interpretation related to neuromuscular control, muscle recruitment, or device-specific effects. These findings highlight substantial inter-individual variability and underscore the need for larger controlled studies to identify predictors of response and optimize rehabilitation protocols. Full article

Background/Objectives: HER2-low breast cancer has emerged as a clinically meaningful category that challenges the historical HER2-positive versus HER2-negative classification. Although not defined as a distinct biological subtype, HER2-low tumors exhibit unique clinicopathological features and differential sensitivity to novel antibody–drug conjugates. Accurate identification remains difficult due to limitations in immunohistochemistry performance, inter-observer variability, intratumoral heterogeneity, and dynamic shifts in HER2 expression over time. This review synthesizes current evidence on the biological and clinical characteristics of HER2-low breast cancer and evaluates emerging diagnostic innovations, with emphasis on liquid biopsy approaches and evolving technologies that may enhance diagnostic accuracy and monitoring. Methods: A narrative literature review was conducted, examining tissue-based HER2 testing, liquid biopsy modalities, including circulating tumor cells, circulating nucleic acids, extracellular vesicles, and soluble HER2 extracellular domains, and applications of artificial intelligence (AI) across histopathology and multimodal diagnostic systems. Results: Liquid biopsy technologies offer minimally invasive, real-time assessment of HER2 dynamics and may overcome fundamental limitations of tissue-based assays. However, these platforms require rigorous analytical validation and face regulatory and standardization challenges before widespread clinical adoption. Concurrently, AI-enhanced histopathology and multimodal diagnostic systems improve reproducibility, refine HER2 classification, and enable more accurate prediction of treatment response. Emerging biosensor- and AI-enabled monitoring frameworks further support continuous disease evaluation. Conclusions: HER2-low breast cancer sits at the intersection of evolving pathology and technological innovation. Integrating liquid biopsy platforms with AI-driven diagnostics has the potential to advance precision stratification and guide personalized therapeutic strategies for this expanding patient subgroup. Full article

Endometrial carcinoma (EC) continues to represent a major cause of gynecologic cancer–related mortality among women worldwide. Its multifactorial etiopathogenesis and underlying molecular heterogeneity have been the focus of extensive investigation. While traditional histological classification provides essential diagnostic insight, it is limited in predicting prognosis and therapeutic response due to significant interobserver variability. Recent advances in molecular biology and cancer genomics have profoundly enhanced understanding of EC pathogenesis. The Cancer Genome Atlas (TCGA) project delineated four distinct molecular subtypes of EC, POLE ultra-mutated, microsatellite instability hypermutated (MSI-H), copy number low (CNL) and copy number high (CNH), each defined by unique genomic alterations, histopathologic features, and clinical behaviors. These molecular groups demonstrate significant prognostic and therapeutic implications, correlating with differential outcomes and treatment responses. This review summarizes current evidence on the genomic landscape of endometrial carcinoma and underscores the pivotal role of molecular classification in improving diagnostic accuracy, prognostic stratification, and personalized therapy. Ongoing research into molecular biomarkers holds promise for refining patient management and optimizing clinical outcomes. Full article

Background/Objectives: Large Language Models (LLMs), including ChatGPT-4 and Gemini, are increasingly incorporated into clinical care; however, their reliability within maternal–fetal medicine (MFM), a high-risk field in which diagnostic and management errors may affect both the pregnant patient and the fetus, remains uncertain. Evaluating the alignment of AI-generated case management recommendations with those of MFM specialists, emphasizing accuracy, agreement, and clinical relevancy. Study Design and Setting: Cross-sectional study with blinded online evaluation (November–December 2024); evaluators were blinded to responder identity (AI vs. human), and case order and response labels were randomized for each evaluator using a computer-generated sequence to reduce order and identification bias. Methods: Twenty hypothetical MFM cases were constructed, allowing standardized presentation of complex scenarios without patient-identifiable data and enabling consistent comparison of AI-generated and human specialist recommendations. Responses were generated by ChatGPT-4, Gemini, and three MFM specialists, then assessed by 22 blinded board-certified MFM evaluators using a 10-point Likert scale. Agreement was measured with Spearman’s rho (ρ) and Cohen’s (κ); accuracy differences were measured with Wilcoxon signed-rank tests. Results: ChatGPT-4 exhibited moderate alignment (mean 6.6 ± 2.95; ρ = 0.408; κ = 0.232, p < 0.001), performing well in routine, guideline-driven scenarios (e.g., term oligohydramnios, well-controlled gestational hypertension, GDMA1). Gemini scored 7.0 ± 2.64, demonstrating effectively no consistent inter-rater agreement (κ = −0.024, p = 0.352), indicating that although mean scores were slightly higher, evaluators varied widely in how they judged individual Gemini responses. No significant difference was found between ChatGPT-4 and clinicians in median accuracy scores (Wilcoxon p = 0.18), while Gemini showed significantly lower accuracy (p < 0.01). Model performance varied primarily by case complexity: agreement was higher in straightforward, guideline-based scenarios and more variable in complex cases, whereas no consistent pattern was observed by gestational age or specific clinical domain across the 20 cases. Conclusions: AI shows promise in routine MFM decision-making but remains constrained in complex cases, where models sometimes under-prioritize maternal–fetal risk trade-offs or incompletely address alternative management pathways, warranting cautious integration into clinical practice. Generalizability is limited by the small number of simulated cases and the use of hypothetical vignettes rather than real-world clinical encounters. Full article