Browse Articles
Discover research articles across all indexed journals
Editorial Expression of Concern: Study on mechanical properties and energy change of rock materials in whole splitting process based on peridynamics
An efficient hybrid quantum machine learning framework for Alzheimer classification
Abstract Neuro-imaging data is necessary for timely and correct diagnosis of the disease Alzheimer’s (AD). Early clinical intervention is challenging due to the high-dimensional nature of the images, making it computationally difficult to work on representations, sparse annotated datasets and inter-subject variance. Deep convolutional VGG, ResNet, DenseNet, MobileNet and other neural networks (CNNs) have shown good performance. Diagnostic performance, but they are based on a high number of parameters and require highly trained personnel. Data can often times restrict resilience to data-skimy clinical settings. To address these challenges, the current paper presents a hybrid Quantum Machine Learning (QML) architecture for Alzheimer’s disease Detection and classification based on a combination of classical CNN-based feature extraction and Variational quantum classifier. In the suggested method, diseased, relevant, and compact features are extracted using pre-trained CNN backbones. Data from structural MRI scans is later translated into quantum states and computed. With parameterised variational quantum circuits optimised with a hybrid quantum-classical learning loop. The proposed method has been experimentally evaluated on benchmark datasets of MRI scans of Alzheimer’s patients. The hybrid CNNQML model has a mean classification accuracy of about 93% which is similar to deep CNN models, including ResNet and DenseNet (94%), and much more. Beating lightweight models like MobileNet and shallow CNNs by six percentage points. It is worth noting that the hybrid QML model performs better in conditions with limited training data. Performance degradation is observed in accuracy and AUC for all CNN variants, with up to 5% lower performance to deep CNNs. Cross-validation, confidence interval, paired significance test, and statistical validation. Effect size analysis proves the fact that the hybrid model produces statistically significant performance, as well as cases of clinically meaningful improvements over lightweight CNN baselines. Similar to the state-of-the-art deep networks with significantly reduced trainable parameters in the classification stage. These findings show that hybrid QML is a data-efficient and powerful substitute in the diagnosis of Alzheimer’s. On the whole, this is a work that sets a practical way to integrate quantum machine learning into clinical neuro-imaging pipelines and hybrid settings. QML has supplementary deep-learning enhancement to classics in next-generation Alzheimer’s Decision-support systems.
Evaluation of different defect-inspection setups for injection molding parts based on the deep learning method
Abstract In the injection molding industry, the shift toward small-batch production has led to a greater variety of products and smaller batch sizes, necessitating frequent mold changes and efficient quality control, which still largely relies on human operators. This study proposes a comprehensive methodology for evaluating and comparing deep learning-based automatic optical inspection (AOI) strategies to detect complex surface defects in injection-molded parts. Three inspection setups were assessed: static frontal imaging, belt conveyor inspection, and robotic-assisted inspection. The findings reveal clear differences in defect detection capabilities among the methods, with the robotic-assisted approach demonstrating superior performance, achieving higher defect detection accuracy due to its flexibility in optimizing camera angles and positions. The proposed methodology serves as a workflow to systematically evaluate and optimize inspection setups across different parameters, enabling informed decisions about AOI systems design. This research contributes to narrowing the gap between the development of advanced detection algorithms and their industrial application, offering insights into the strategic implementation of AI technologies in quality control processes and enhancing the automatic detection of challenging defects.
Task-moment mismatch: diagnosing failures of coverage-based methods on heavy-tailed distributions in coreset selection
Abstract Coreset construction aims to identify a small, weighted subset of training data that approximates the full dataset’s statistical properties, thereby enabling efficient model training without significant performance loss. Existing coreset selection methods typically optimize for geometric coverage or mean matching, under the implicit assumption that these objectives transfer meaningfully to arbitrary downstream tasks. In this study, it is demonstrated that this assumption can fail when the downstream task depends on statistical moments that the selection objective often does not preserve. A common underlying phenomenon termed the task-moment mismatch is identified, wherein coreset objectives that neglect task-relevant statistical moments yield worse performance than uniform random sampling. The objective of this work is threefold: (1) to diagnose failure modes of coverage-based coreset selection on heavy-tailed distributions, (2) to characterize how optimizing for lower-order moments can degrade the preservation of higher-order statistical structure required by certain tasks, and (3) to provide a constructive demonstration that moment-aware selection can address this mismatch. A greedy selection procedure termed Hierarchical Moment-Preserving (HMP) selection is proposed, which targets specific moment tensors through a two-stage hierarchical approach. Experimental evaluation across synthetic distributions and realistic signals parameterized from real-world sources reveals that k-means++ exhibits $$5\times$$ higher covariance error than random sampling on heavy-tailed data, confirming that coverage optimization can be counterproductive for moment-dependent tasks. Furthermore, covariance-optimized coresets are shown to cause $$-13.3$$ dB signal separation degradation on Independent Component Analysis tasks ( $$p = 0.002$$ ), attributable to the disruption of fourth-order cumulant structure that ICA requires. The proposed moment-aware approach achieves $$3\times$$ lower covariance error than random sampling and $$+15\%$$ improvement in outlier detection precision on financial data. A practical decision matrix is provided to guide practitioners in matching coreset methods to the statistical requirements of their downstream tasks.
Novel gas tungsten powder filler metal arc welding process
Abstract In the present investigation, an innovative and patented welding process referred to as gas tungsten powder filler metal arc welding (GTPFAW) has been developed, utilizing a micro powder mixture as a filler metal in lieu of conventional welding wires. This approach presents numerous advantages, encompassing a refined welding microstructure, enhanced mechanical properties, adaptability in composition tailored for various welding applications, and the elimination of the necessity for utilizing prefabricated welding wires. Steel grade S275 JR was employed as the foundational metal to be joined with the powder filler metal, which possessed a composition corresponding to that of the standard solid wire ER 70-S3. The evaluation of the weldability of gas tungsten powder filler metal arc welding (GTPFAW) methodology was conducted through a comprehensive analysis of the overall joint integrity and its corresponding mechanical attributes. The findings indicated that GTPFAW exhibited user-friendliness and yielded effective welding outcomes. The joint properties that resulted from the novel approach of GTPFAW were methodically contrasted with the joint properties generated by the traditional gas tungsten arc welding (GTAW) technique in combination with solid wire ER 70-S3, ensuring equivalent joint dimensions and the same steel parent metal classification S275 JR. The microstructural examination disclosed a significant acicular ferrite morphology characterized by its ultra-fine nature, markedly finer than the conventional microstructure produced by the solid filler. Furthermore, the weld metal produced exhibited a consistent hardness range (135–143 HV5) throughout its entire cross-section. This innovative technique possesses the capacity to enable a broad spectrum for the enhancement of filler metal formulations, thus providing substantial opportunities for the union of metals that display or possess particular or distinctive compositions. Moreover, this novel welding approach may be regarded as an environmentally sustainable practice and a proponent of the 2030 Agenda for Sustainable Development.
Closing the sepsis gap: from molecular mechanisms to scalable bedside care
Abstract Recent estimates point to roughly 49 million cases and 11–13 million deaths annually due to sepsis, with a disproportionate burden borne by low- and middle-income countries. Three decades of progress, summarised in the recently updated Surviving Sepsis Campaign guidelines, have refined how clinicians screen for, resuscitate, and treat patients with sepsis. Yet, persistent uncertainty surrounds early identification, pathogen detection, hemodynamic targets, adjunctive therapies, and the long-term burden carried by survivors. This Collection, Sepsis: Treatment, intervention, mortality, brings together original research that maps where the field is moving: pragmatic diagnostics that run on standard hospital equipment; multi-omic biomarker discovery in blood, plasma extracellular vesicles, and urine; computational dissection of pathogen biology and longitudinal host-response trajectories; preclinical interrogation of metabolic and signalling pathways implicated in organ injury; and mechanistic studies of the muscle and mitochondrial wasting that shape life after sepsis. Together, these papers sketch a research agenda for sepsis that is simultaneously molecularly precise and operationally scalable.
Development and validation of a food frequency questionnaire for caffeine intake and dietary assessment in Iranian adolescents
Deep learning-based first-trimester down syndrome risk prediction in East Asian pregnant women using simplified clinical indicators
Reconfigurable hybrid beamforming for 6G wireless systems across quasi-far-field and strong near-field regimes
Synthetic data to boost under-represented patients and create virtual trial cohorts: the RATE-AF case study
Abstract Clinical trials are essential for medical progress, but in certain circumstances can be constrained by recruitment costs, ethical challenges, and limited diversity in participant representation, reducing the generalisability of findings across all subgroups. In these cases, the development of digital approaches that complement traditional trials could be of particular value. We present a Bayesian network framework for synthetic data generation in clinical trials, designed to (1) boost the representation of small and under-represented subgroups and (2) generate virtual patient cohorts that replicate full trial populations with high fidelity. The framework combines probabilistic modelling with conditional synthetic data generation and is evaluated using data from the RAte control Therapy Evaluation in permanent Atrial Fibrillation (RATE-AF) randomised controlled trial, a study that compared two treatments for rate control (digoxin versus bisoprolol, a beta-blocker) in patients with atrial fibrillation and symptoms of heart failure. The framework was first assessed in a controlled boosting experiment, designed to recover simulated subgroup under-representation within the original cohort, and then extended to an exploratory boosting scenario to examine hypothetical increases in subgroup representation, before being applied to replicate the full trial population. Across these settings, it preserved statistical fidelity and reproduced the analytical results observed in the real data, boosting under-represented subgroups where sufficient data are available, whilst acknowledging limitations of boosting under extreme small-sample scenarios. This study positions synthetic data as a potential digital pathway that, with further development, could be used to support real-world clinical trials where recruitment of some population subgroups may be challenging.
Simulation-based state and health co-estimation for battery–supercapacitor hybrid energy storage systems
Diatomaceous earth as physical control agent for root-knot nematode
Abstract This study evaluated diatomaceous earth (DE), a naturally occurring mineral composed primarily of amorphous silicon dioxide (SiO₂), as a physical control agent against the root-knot nematode Meloidogyne javanica infecting eggplant ( Solanum melongena ). Under laboratory conditions, DE was tested against second-stage juveniles (J2) at different concentrations. Among the concentrations of 0.01, 0.05, 0.1, 0.3, 0.5, 1.0, and 2.0 g/L, mortality rates of 100% were observed at concentrations ranging from 0.05 to 2.0 g/L, while 85% mortality was recorded at the lowest concentration (0.01 g/L). In greenhouse conditions, DE was applied at concentrations of 0.05, 0.5, 1.0, and 2.0 g/L using three application methods: root dipping and foliar spray at inoculation time, and soil drench at three distinct time points relative to nematode inoculation. All treatments significantly ( P ≤ 0.05) reduced nematode infection parameters in a concentration-dependent manner. Root dipping at 2.0 g/L resulted in reductions of 92% (galls), 93% (egg masses), and 92% (adult females). Soil drench applied at the time of inoculation produced reductions of 86%, 85%, and 85% for the same parameters, respectively; while foliar spray resulted in reductions of 73%, 68%, and 73%. Among the application methods, soil drench consistently produced the greatest suppression of J2 populations. In addition to nematode control, DE application was associated with increased plant growth parameters. Soil drench applied one week after inoculation increased root dry weight by 178% at 0.5 g/L and by 185% at 2.0 g/L. These results indicate that DE has potential as a physical control agent against M. javanica infecting eggplant and may warrant further investigation for integration into nematode management strategies.
DASTNet-X: an explainable dual-attention based spatio-temporal network for arrhythmia classification
A DCN+CSF1+ fibroblast subpopulation drives pathological fibrosis in ARDS by exploiting the TGF-β signaling axis
Asymmetric dual pathway fusion for histology driven spatial transcriptomic prediction
A numerical simulation study on dual-mineral carbonate rocks incorporating thermo-hydro-chemical (THC) coupling
A multiplicative additive bias variational framework for accurate and interpretable brain MRI segmentation in cloud-based medical imaging systems
Abstract Accurate brain magnetic resonance imaging (MRI) segmentation remains challenging due to intensity inhomogeneity, acquisition-related bias fields, and ambiguous tissue boundaries. To address these challenges, a Multiplicative–Additive Bias Single-Function Dual-Level-Set (MAB-SFDLS) model is introduced within a Software-as-a-Service (SaaS)-based medical image analysis framework. The model incorporates both multiplicative and additive bias components into a unified variational energy formulation and employs a single level-set function with dual thresholds to achieve stable multi-region segmentation with smooth and continuous boundaries. The method was evaluated on the MRBrainS18 dataset, achieving Dice scores of 0.95 for white matter and 0.86 for gray matter, with a boundary deviation of 2.20 mm measured using HD95. Compared with the classical level-set formulation, notable improvements were observed in both overlap accuracy and boundary precision. The approach also demonstrated competitive performance against state-of-the-art deep learning models, including nnU-Net and U-Mamba, while maintaining lower computational requirements. Statistical analysis confirmed that the improvements were significant (p < 0.05). To enhance interpretability and practical applicability, the segmentation framework is integrated with a browser-based 3D visualization module that supports synchronized surface and volume rendering, as well as interactive region-of-interest exploration. This framework provides a practical, interpretable, computationally efficient, and scalable approach to robust brain MRI segmentation in a cloud-based medical imaging environment. The proposed model code and SaaS platform prototype are publicly available at https://doi.org/10.5281/zenodo.20797546 .
The role of a confining bulk phase in modulating activity and reorganization within viscous membranes
Abstract Active molecular motors coordinate their action to reorganize on fluid membranes to drive faster or more efficient cellular transport. The resulting membrane hydrodynamics have been shown to enhance in-plane transport and guide self-assembly. In this work, we examine the qualitative differences that arise in the collective motion of active, force-free inclusions as a result of ‘confining’ the extent of the 3D fluid phase adjacent to the membrane. A thin film underlying the membrane screens hydrodynamic interactions and modifies the effective Saffman-Delbrück length. We characterize resulting features in cluster formation and lipid-field reorganization using large-scale simulations. We also analytically investigate the stability of typical aggregate structures seen in simulations, demonstrating the role of hydrodynamic interactions modulated by the extent of bulk phase in driving aggregation and reorganization in the plane of the membrane. Put together, we reveal new mechanisms of driving lipid transport and coordinated motion by modulating collective activity within membranes.