Browse Articles

‘Megacluster’ of genes enables bacteria to make potent antibiotic mixture

Low-speed UAV trajectory prediction under short-term data loss: a preliminary feasibility study for vehicle-mounted electro-optical tracking

An exploratory SMOTE-SVM approach for identifying preoperative biomechanical risk factors driving early toric intraocular lens micro-rotation in extremely imbalanced cohorts

Abstract The predictive modeling of postoperative mechanical complications, such as the early micro-rotation of premium toric intraocular lenses (IOLs), is severely hindered by the extreme imbalance of clinical datasets. Traditional statistical methods often fail to capture complex biomechanical interactions in rare-event scenarios. This exploratory pilot study introduces a machine learning framework designed as a hypothesis-generating tool to handle extremely imbalanced ophthalmic data and identify potential preoperative biometric features associated with toric IOL micro-rotation. A prospective cohort of 35 eyes implanted with the Clareon PanOptix® Toric IOL was analyzed, quantifying true rotational stability via high-resolution photographic registration. Given the exceedingly low incidence of > 1-degree micro-rotation, a strict, leak-proof fivefold cross-validation pipeline was established. The Synthetic Minority Over-sampling Technique (SMOTE) was applied exclusively within the training folds, and an interpretable Linear Support Vector Machine (Linear SVM) was deployed to extract robust feature weights for biomechanical interpretation. Our findings highlight the “accuracy paradox” in small clinical datasets: complex ensemble models exhibited severe majority-class bias, failing to detect rare micro-rotations. Conversely, the SMOTE-enhanced Linear SVM achieved a Precision-Recall Area Under the Curve (PR-AUC) of 0.463, outperforming a random baseline by nearly a factor of three. The algorithmic feature weights successfully isolated Anterior Chamber Depth (ACD) and steep keratometry (Steep K2) as the primary geometric drivers of rotational instability, demonstrating a profound alignment with clinical ocular biomechanics. While strictly constrained by the small sample size (N = 35) and limited event rate, this preliminary pilot framework successfully bridges high-dimensional data augmentation with physical ocular biomechanics, effectively identifying minority risk features and laying the groundwork for future AI-driven surgical navigation systems.

Prediagnostic opioid use and survival in multiple myeloma: a nationwide register-based study

Abstract Multiple myeloma (MM) accounts for approximately 1% of incident cancers worldwide and carries substantial mortality. Diagnostic delay is common, with more than one-third of patients requiring emergency care at diagnosis, a presentation linked to inferior prognosis. Pain is a frequent early MM symptom, yet the extent and prognostic relevance of prediagnostic opioid use remains poorly characterized. Using nationwide Danish registers, we conducted a matched case-control analysis of prediagnostic opioid use, irrespective of indication, and its association with overall survival (OS) among MM patients. The study included 6,953 MM patients diagnosed in 2001–2021 and 101,793 matched population controls. Compared with controls, opioid use was first elevated among MM cases approximately 2.25 years before diagnosis, increasing more consistently closer to diagnosis. High prediagnostic opioid use (≥ 10 mg daily oral morphine equivalents) was associated with shorter OS, compared with no opioid use (1-year OS 72% vs. 82%; adjusted hazard ratio 1.23, 95% CI 1.07–1.41). These findings indicate that opioid use increases closer to MM diagnosis date, likely reflecting pain related to undiagnosed MM. Furthermore, prediagnostic opioid use was associated with shorter OS independently of available prognostic factors, including ISS stage and myeloma bone disease.

Physics-inspired enhancement framework for industrial digital radiography based on radiation–matter interaction modeling

Progress in wheat farm yield improvement has largely stagnated across the U.S. Pacific Northwest

Abstract Yield stagnation, which has occurred in some global wheat production hubs, is a serious barrier to meeting future food demands. Yield trends across the U.S. Pacific Northwest, where 8.2 million Mg of wheat was produced in 2025, are mostly unknown. Using publicly-available data (USDA farm surveys and university-led variety trials), the objective of this research was to evaluate winter wheat yield trends across Oregon, Washington, and Idaho since the early 1970s. Survey analysis showed that historic rates of yield increase among agricultural districts ranged from 18 to 195 kg ha −1  year −1 , with incremental shifts toward irrigation in select districts driving the highest rates. But yields began to stagnate between the early 1980s and mid-1990s. Analysis of top-yielding varieties in 163 variety trials conducted since 2000 showed that genetic gain by year of variety release has slowed to 4–5 kg ha −1  year −1 across lower- and higher-yielding environments. Variety trial yield trends by test year suggest that environmental and/or agronomic factors may be negatively offsetting genetic gains, especially in higher-yielding environments. These findings have broad implications for regional farmers and multidisciplinary researchers who support them. Gaining more rigorous understanding of the factors contributing to yield stagnation is critical to reversing the trend.

Genomic insights into the population dynamics and demise of Neanderthals

Adaptive cellular evolution in the intestine of hyperdiverse cichlid fishes

Design and evaluation of an intent-based web of things query framework for smart device discovery

Detection of anisotropic cosmic structures on a gigaparsec scale

Ontogenetic emergence of behavioral consistency in a self-fertilizing fish

A deep learning model to predict objective response to TACE and TKI-based therapy in HBV-related uHCC

Genetic diversity of late Neanderthals in northwestern Europe

Abstract Archaeological, osteological and genetic evidence suggests that Neanderthals lived in small groups 1,2 ; however, less is known about whether these groups were part of isolated communities or belonged to larger, well-connected populations 3 . The dense concentration of broadly contemporaneous Neanderthal sites in the Meuse Basin, Belgium 4 , provides a rare opportunity to study regional populations at high resolution. Here we generated genetic data from 27 Neanderthals who lived less than approximately 52,500 years ago from ten archaeological sites in Belgium and France, including a high-coverage genome from a 45,000-year-old individual from Goyet, Belgium. We show that most of these individuals are more closely related to one another than to other contemporaneous late Neanderthals in Europe. Further, some of these individuals carry DNA from a Neanderthal lineage predating the split of late Neanderthals. Although these Neanderthals overlapped temporally with early modern humans in northwestern Europe from around 47,000 years ago, we find no evidence of recent gene flow from modern humans. They also do not show the genetic signatures of mating among close relatives found in Altai Neanderthals, suggesting that they lived in larger or better-connected groups. Moreover, genetic load did not accumulate over time, arguing against progressive genetic deterioration as a driver of Neanderthal extinction.

ACADPro: XAI-student procrastination classification in academia using optuna optimized machine learning models

Adaptive user clustering enhanced BiLSTM-attention for short-term load forecasting in smart distribution networks

Transformer-based classification with enhanced causal explainability from otoscopic images

Abstract Otitis media is a major cause of hearing loss, particularly in children. However, nonspecific symptoms and subjective evaluations make its diagnosis challenging. To address this, we developed transformer-based models to classify tympanic membrane conditions from otoscopic images. This approach aims to enhance diagnostic transparency and reliability in clinical settings. We trained vision transformer (ViT) and Data-efficient Image Transformer (DeiT) models on 454 pediatric and adult otoscopic images. These models performed multi-class classification to distinguish between normal, effusion, and tube conditions. For explainability, we utilized Gradient-weighted Class Activation Map (Grad-CAM), Layer-wise Relevance Propagation (LRP), and Attention Rollout (AR). Furthermore, we introduced a hybrid fusion strategy based on Canonical Correlation Analysis. The framework’s effectiveness was then evaluated using insertion and deletion causal metrics. The ViT model achieved an accuracy of 97.78% (AUC: 0.998), outperforming DeiT, which reached 93.33% (AUC: 0.994). Notably, ViT attained an F1-score of 97.30% for the effusion class. Among the Explainable Artificial Intelligence (AI) methods, the hybrid LRP and AR approach provided the highest explainability. It yielded an average deletion score of 0.3008 and an insertion score of 0.8918, precisely highlighting critical image features for model predictions. In conclusion, integrating transformer-based models with hybrid explainability methods significantly enhances diagnostic transparency. These advancements foster clinician trust and lay a strong foundation for reliable clinical decision support systems.

Assessment of visual performance and colour vision awareness among graphic designers in Ghana

Sex-specific CT-derived reference cutoffs for body composition in healthy Brazilian adults: a multicenter study

In silico design of novel CTL based multi epitope vaccine for esophageal cancer using immunoinformatics and molecular docking

Abstract Esophageal cancer is an aggressive malignancy with high morbidity, mortality, and limited durable treatment options due to tumor heterogeneity, immune evasion, and recurrence. This study addresses these challenges by computationally designing a novel CTL-based multi-epitope vaccine using experimentally validated epitopes from cancer-testis antigens (NY-ESO-1 and MAGE-A family), which are overexpressed in esophageal squamous cell carcinoma. To the best of our knowledge, this represents one of the most comprehensive in silico investigations for esophageal cancer, uniquely integrating experimentally validated epitopes with advanced immunoinformatics, high-resolution structural modeling, molecular dynamics, and immune simulation strategies. Nine experimentally validated CTL epitopes were retrieved from IEDB and rigorously evaluated for antigenicity (VaxiJen), toxicity (ToxinPred), allergenicity (AllerTOP), and IFN-γ induction (IFNepitope). A 253-amino-acid multi-epitope construct was assembled with AAY/EAAAK linkers, PADRE adjuvant, and 5 S rRNA-derived TLR4 agonist. Physicochemical properties were assessed (ProtParam, SOLpro); secondary/tertiary structures predicted (SOPMA, trRosetta); and validated (ProSA, Ramachandran). B-cell epitopes were predicted with ElliPro. Molecular docking (ClusPro) with TLR4, 100-ns MD simulations (GROMACS), and MM/GBSA binding free energy calculations were performed. Immune responses were simulated using C-ImmSim, and population coverage was analyzed via IEDB. The vaccine construct demonstrated excellent stability (instability index 31.16), solubility (0.577), and antigenicity (VaxiJen 0.5734; non-allergenic). It exhibited a predominantly α-helical structure (64.43%) with high model quality (ProSA Z-score: − 6.33). Strong TLR4 binding was confirmed (–910.7 kJ/mol, stable RMSD ~ 0.29 nm, MM/GBSA − 110.76 kcal/mol). Immune simulations predicted robust IgG/IgM responses, memory cell formation, and elevated IFN-γ (> 4 × 10⁵ ng/mL). Global population coverage reached 50.02%. This novel CTL-based multi-epitope vaccine candidate is stable, immunogenic, and capable of eliciting strong anti-tumor immunity. It provides a promising computational platform for esophageal cancer immunotherapy, warranting experimental validation and clinical translation.

Optical cooling by interfacial charge transfer in 2D heterostructures