Search Results (6,639)

This study presents Care-MOVE, a smartphone-based application designed for continuous, passive, and unobtrusive monitoring of mobility, environmental exposure, and cognitive status in older adults within a telemedicine framework. The system integrates movement-related data collected through smartphone sensors (GPS, activity recognition, and caloric expenditure estimation) with contextual air quality information and standardized neuropsychological assessments, resulting in a comprehensive multimodal dataset (Care-MOVE Dataset). An exploratory proof-of-concept study was conducted on a subsample of 53 participants aged over 65, each monitored continuously for five days, contributing on average more than 30,000 longitudinal records. To investigate whether daily motor behavior can serve as a digital biomarker of cognitive functioning, several Machine Learning and Deep Learning models were evaluated using a Leave-One-User-Out (LOUO) cross-validation strategy. The comparative analysis included traditional classifiers (Logistic Regression, Random Forest, Gradient Boosting, K-Nearest Neighbors, and Support Vector Machines) as well as temporal deep learning architectures (1D CNN, LSTM, GRU, and Transformer). Among all of the evaluated approaches, the Support Vector Machine with RBF kernel achieved the best performance, reaching an accuracy of 98.1%, a balanced accuracy of 0.988, and an F1-score of 0.981, demonstrating robust generalization across unseen subjects. For this reason, the study was designed and presented as an exploratory proof-of-concept rather than a definitive clinical validation. This integrated approach not only enables the collection of detailed and contextualized data but also opens new perspectives for proactive digital healthcare, focused on risk prevention, improving quality of life, and promoting autonomy in elderly patients. Full article

Background: Rapid and accurate fracture detection is critical in emergency departments (EDs), where high patient volume and time pressure increase the risk of diagnostic error, particularly in radiographic interpretation. Multimodal large language models (LLMs) with image-recognition capability have recently emerged as general-purpose tools for clinical decision support, but their diagnostic performance within routine emergency department imaging workflows in orthopedic trauma remains unclear. Methods: In this retrospective diagnostic accuracy study, we included 1136 consecutive patients referred from the ED to orthopedics between 1 January and 1 June 2025 at a single tertiary center. Given the single-center, retrospective design, the findings should be interpreted as hypothesis-generating and may not be fully generalizable to other institutions. Emergency radiographs and clinical data were processed by a multimodal LLM (2025 version) via an official API using a standardized, deterministic prompt. The model’s outputs (“Fracture present”, “No fracture”, or “Uncertain”) were compared with final diagnoses established by blinded orthopedic specialists, which served as the reference standard. Diagnostic agreement was analyzed using Cohen’s kappa (κ), sensitivity, specificity, accuracy, and 95% confidence intervals (CIs). False-negative (FN) cases were defined as instances where the LLM reported “no acute fracture” but the specialist identified a fracture. The evaluated system is a general-purpose multimodal LLM and was not trained specifically on orthopedic radiographs. Results: Overall, the LLM showed good diagnostic agreement with orthopedic specialists, with concordant results in 808 of 1136 patients (71.1%; κ = 0.634; 95% CI: 68.4–73.7). The model achieved balanced performance with sensitivity of 76.9% and specificity of 66.8%. The highest agreement was observed in knee trauma (91.7%), followed by wrist (78.8%) and hand (69.6%). False-negative cases accounted for 184 patients (16.2% of the total cohort), representing 32.4% of all LLM-negative assessments. Most FN fractures were non-displaced (82.6%), and 17.4% of FN cases required surgical treatment. Ankle and foot regions showed the highest FN rates (30.4% and 17.4%, respectively), reflecting the anatomical and radiographic complexity of these areas. Positive predictive value (PPV) and negative predictive value (NPV) were 69.4% and 74.5%, respectively, with likelihood ratios indicating moderate shifts in post-test probability. Conclusions: In an emergency department-to-orthopedics consultation cohort reflecting routine clinical workflow, a multimodal LLM demonstrated moderate-to-good diagnostic agreement with orthopedic specialists, broadly within the range reported in prior fracture-detection AI studies; however, these comparisons are indirect because model architectures, training strategies, datasets, and endpoints differ across studies. However, its limited ability to detect non-displaced fractures—especially in anatomically complex regions like the ankle and foot—carries direct patient safety implications and confirms that specialist review remains indispensable. At present, such models may be explored as hypothesis-generating triage or decision-support tools, with mandatory specialist confirmation, rather than as standalone diagnostic systems. Prospective, multi-center studies using high-resolution imaging and anatomically optimized algorithms are needed before routine clinical adoption in emergency care. Full article

Renewable Energy Communities are expected to play a key role in the decarbonization of power systems, but their design and operation involve multiple, often conflicting objectives and evolving regulatory frameworks. However, prospective REC promoters and members must make early-stage design choices under policy constraints while balancing economic, environmental, and reliability goals, which motivates the need for transparent and reproducible decision-support tools. This paper presents Adapters, a two-level decision-making tool that couples long-term planning with short-term operational adaptation for hybrid renewable energy systems. The core optimisation model is explicitly multi-objective, with three weighted terms (w₁, w₂, and w₃) that represent total cost, CO₂ emissions, and unserved energy, respectively, allowing users to explore trade-offs between economic performance, environmental impact, and reliability. The tool integrates detailed component models (such as photovoltaic, wind, and battery storage) with a flexible optimisation layer and architecture compatible with digital-twin approaches. Its capabilities are illustrated through prototype single-household case studies, showing how different stakeholder preferences and regulatory conditions can be reflected in the choice of objective weights and system configurations. The overall aim is to provide a transparent and reproducible environment to support the emergence and operation of RECs in line with EU energy and climate goals. Full article

Impinging jet ventilation (IJV) systems have attracted significant attention due to their potential to augment indoor thermal comfort and airflow distribution. Previous studies have primarily investigated corner and mid-wall IJV installations; however, comparative analyses focusing on different diffuser geometries remain limited. Accordingly, this study examines the combined effects of IJV diffuser geometry and installation location on thermal comfort indices and airflow characteristics. A full three-dimensional computational fluid dynamics (CFD) model, without the use of symmetry, is developed to simulate a realistic office environment (3 × 3 × 2.9 m³), operating in cooling mode under hot summer climatic conditions. Three IJV diffuser cross-section geometries—triangular, square, and circular—are evaluated at four installation locations (two corners and two mid-wall positions), assuming a fixed occupant location. A combined return and exhaust outlet configuration is adopted. The results indicate that the IJV location influences airflow and temperature distributions more strongly than the diffuser geometry. Nevertheless, the circular diffuser exhibits superior performance compared to the triangular and square geometries. The mid-wall location placed behind the occupant and away from the hot exterior wall demonstrates reduced thermal stratification, improved airflow characteristics, and weaker vortex formation, making it the most favorable configuration. From an architectural perspective, these findings highlight the importance of early coordination between ventilation design and office spatial planning, as diffuser placement directly influences occupant comfort zones and furniture layout. Moreover, the preference for mid-wall installations supports a more flexible façade design and allows for greater freedom in organizing workspaces without compromising thermal performance. Full article

In recent years, the rapid commercialization and widespread adoption of portable and wearable electronic devices have imposed increasingly stringent performance requirements on flexible sensors, including enhanced sensitivity, stability, response speed, comfort, and integration. This trend has driven extensive research and technological advancement in sensor material systems, among which carbon-based materials have emerged as core candidates for high-performance flexible sensors due to their exceptional electrical conductivity, mechanical flexibility, chemical stability, and highly tunable structural features. Meanwhile, new sensing mechanisms and innovative device architectures continue to emerge, demonstrating significant value in real-time health monitoring, early disease detection, and motion-state analysis, thereby expanding the functional boundaries of flexible sensors in the health-care sector. This review focuses on the application progress and future opportunities of carbon-based materials in flexible sensors, systematically summarizing the critical roles and performance-optimization strategies of carbon nanotubes, graphene, carbon fibers, carbon black, and their derivative composites in various sensing systems, including strain and pressure sensing, physiological electrical signal detection, temperature monitoring, and chemical or environmental sensing. In response to the growing demands of modern health-monitoring technologies, this review also examines the practical applications and challenges of flexible sensors—particularly those based on emerging mechanisms and novel structural designs—in areas such as heart-rate tracking, blood-pressure estimation, respiratory monitoring, sweat-component analysis, and epidermal electrophysiological signal acquisition. By synthesizing the current research landscape, technological pathways, and emerging opportunities of carbon-based materials in flexible sensors, and by evaluating the design principles and practical performance of diverse health-monitoring devices, this review aims to provide meaningful reference insights for researchers and support the continued innovation and practical deployment of next-generation flexible sensing technologies. Full article

The categorization of real-time defects in heterogeneous domains is a long-standing challenge in the field of industrial visual inspection systems, primarily due to significant visual variations and the lack of labelled information in real-world inspection settings. This work presents the Adaptive Attention DropBlock (AADB) framework, a lightweight deep learning framework that was developed to promote cross-domain defect detection using attention-guided regularization. The proposed architecture integrates the Convolutional Block Attention Module (CBAM) and an organized DropBlock-based regularization scheme, creating a unified and robust framework. Although CBAM-based approaches improve localization of defect-related areas and traditional DropBlock provides a generic spatial regularization, neither of them alone is specifically designed to reduce domain overfitting. To address this limitation, AADB combines attention-directed feature refinement with a progressive, transfer-aware dropout policy that promotes the learning of domain-invariant representations. The proposed model is built on a MobileNetV2 base and trained through a two-phase transfer learning regime, where the first phase consists of pretraining on a source domain and the second phase consists of adaptation to a visually dissimilar target domain with constrained supervision. The overall analysis of a metal surface defect dataset (source domain) and an aircraft surface defect dataset (target domain) shows that AADB outperforms CBAM-only, DropBlock-only, and conventional MobileNetV2 models, with an overall accuracy of 91.06%, a macro-F1 of 0.912, and a Cohen’s k of 0.866. Improved feature separability and localization of error are further described by qualitative analyses using Principal Component Analysis (PCA) and Grad-CAM. Overall, the framework provides a practical, interpretable, and edge-deployable solution to the classification of cross-domain defects in the industrial inspection setting. Full article

Accurate lung tumor segmentation using computed tomography (CT) scans is needed for efficient tumor treatment. However, the development of deep learning models is often constrained by strict patient privacy regulations that limit direct data sharing. This work presents a system that enables multi-institutional collaboration while training high-quality lung tumor segmentation models without requiring access to sensitive patient data. The proposed framework features the AuraViT suite, which includes the standard AuraViT—a hybrid model with 136 million parameters that combines a Vision Transformer (ViT) encoder, Atrous Spatial Pyramid Pooling (ASPP), and attention-gated residual connections—and the Lightweight AuraViT (LAURA) family (Small, Tiny, and Mobile). These variants are designed for resource-constrained environments and potential edge deployment scenarios. Training is conducted on publicly available datasets (MSD Lung and NSCLC) in a simulated five-client federated learning setup that emulates collaboration among institutions while ensuring patient privacy. The framework uses a federated learning setup with FedProx, adaptive weighted aggregation, and a dynamic virtual client strategy to handle data and system differences. The framework is further evaluated through ablation studies on model architecture and feature importance. The results show that the standard AuraViT-FL achieves a global mean Dice score of 80.81%, while maintaining performance close to centralized training. Additionally, the LAURA variations show a better trade-off between accuracy and efficiency. Notably, the Mobile variant with ∼5 M parameters reduces model complexity by over 96% while maintaining competitive performance (82.96% Dice on MSD Lung). Full article

The evaluation of research and development (R&D) project proposals plays a critical role in shaping national scientific and technological priorities. However, existing expert review systems are often characterized by fragmented digital workflows, limited traceability of decisions, and a strong reliance on manual coordination, which reduces transparency and auditability. This paper proposes a conceptual and methodological framework for a national research project evaluation system that integrates artificial intelligence and blockchain technologies as complementary decision-support and data integrity mechanisms. The framework formalizes the complete evaluation lifecycle, including applicant authorization, formal compliance verification, originality and plagiarism analysis, expert selection and assessment, analytical consolidation of reviews, and fixation of final decisions. Artificial intelligence modules are introduced to support thematic classification, compliance checking, expert matching, and analytical processing of expert evaluations, while blockchain technology is incorporated as an immutable integrity layer for recording critical evaluation events and ensuring data provenance. The proposed approach focuses on architectural design, governance principles, and process modeling rather than system implementation or empirical validation. The framework is intended to serve as a reference model for the design and future development of transparent, accountable, and scalable research project evaluation platforms at national and institutional levels. Full article

The increasing frequency of interior renovation and fit-out in office buildings raises concerns about indoor environmental quality, occupant health, and sustainability performance, yet existing certification systems remain largely design-stage or whole-building oriented and provide limited guidance for recurring renovation cycles. This study develops a health-oriented assessment framework for office interior renovation as a structured decision-support tool for practitioners and policymakers. We adopted an integrated approach combining a targeted literature review, expert consultation, the Fuzzy Delphi Method (FDM) for indicator screening, and the Analytic Hierarchy Process (AHP) for hierarchical weighting, based on an expert panel of 20 professionals spanning green building certification, architecture/interior design, MEP engineering, property/facility management, and energy/environmental consulting. Through consensus screening and weighting, four assessment dimensions and eighteen key indicators were identified and prioritized. Environmental quality was ranked highest (39.2%), followed by safety management (23.0%), functional usability (21.1%), and resource efficiency and circularity (16.7%). At the indicator level, indoor air quality management, Heating, Ventilation and Air Conditioning (HVAC) energy efficiency, space-friendly layout, preliminary assessment and planning, and thermal comfort emerged as the top priorities. Overall, the framework bridges the gap between certification-oriented evaluation and the operational realities of office renovation, enabling more consistent integration of health and sustainability considerations across renovation decision-making. Full article

Ultrasound imaging is widely used for early breast cancer screening to enhance patient survival. However, interpreting these images is inherently challenging due to speckle noise, low lesion-to-tissue contrast, and highly variable tumor morphology within complex anatomical structures. Additionally, variations in image characteristics across institutions and devices further impede the development of robust and generalizable computer-aided diagnostic systems. To alleviate these issues, this paper presents a cross-domain segmentation prior guided classification strategy for robust breast tumor diagnosis in ultrasound imaging, implemented through a novel Dual-Stream Diagnosis Network (DSDNet). DSDNet adopts a decoupled dual-stream architecture, where a frozen segmentation branch supplies spatial priors to guide the classification backbone. This design enables stable and accurate performance across diverse imaging conditions and clinical settings. To realize the proposed DSDNet framework, three novel modules are created. The Dual-Stream Mask Attention (DSMA) module enhances lesion priors by jointly modeling foreground and background cues. The Segmentation Prior Guidance Fusion (SPGF) module integrates multi-scale priors into the classification backbone using cross-domain spatial cues, improving tumor morphology representation. The Mamba-Inspired Linear Transformer (MILT) block, built upon the Mamba-Inspired Linear Attention (MILA) mechanism, serves as an efficient attention-based feature extractor. On the BUSI, BUS, and GDPH_SYSUCC datasets, DSDNet achieves ACC values of 0.878, 0.836, and 0.882, and Recall scores of 0.866, 0.789, and 0.878, respectively. These results highlight the effectiveness and strong classification performance of our method in ultrasound breast cancer diagnosis. Full article

Background/Objectives: Accurate nuclear segmentation is a fundamental step in computational pathology, enabling reliable estimation of cellularity and nuclear morphology. However, segmentation models are typically evaluated under ideal imaging conditions, while real-world microscopy data are affected by staining variability, noise, and image degradation. This study aims to comparatively evaluate three representative convolutional architectures for nuclei segmentation, with emphasis on robustness and clinical relevance under perturbed imaging conditions. Methods: U-Net, Attention U-Net, and U-Net++ were trained and evaluated on the BBBC038 nuclei microscopy dataset using fixed train–validation–test splits. Robustness was assessed under three types of synthetic perturbations: Gaussian blur, additive noise, and color jitter. Segmentation performance was quantified using the Dice coefficient and Intersection-over-Union (IoU). Paired Wilcoxon signed-rank tests with Holm correction and Cliff’s delta were used for statistical comparison. In addition, clinically relevant nuclear descriptors—nuclear count, median nuclear area, area interquartile range (IQR), and nuclear density—were extracted from predicted masks, and descriptor stability was analyzed as relative deviation from clean conditions. Results: Under clean imaging conditions, Attention U-Net achieved the highest mean Dice score, while paired statistical analysis indicated that U-Net++ exhibited the most consistent performance across test samples. Under image perturbations, Attention U-Net demonstrated greater robustness to blur and noise, whereas U-Net++ showed superior stability under color variations. Descriptor-based analysis further indicated that U-Net++ preserved nuclear count and density most reliably under chromatic perturbations, while U-Net exhibited larger instability in nuclear count and density, particularly under noise. Conclusions: Architectural design choices strongly influence not only pixel-level segmentation accuracy but also the stability of clinically relevant nuclear morphology descriptors. Robustness evaluation under multiple perturbation types reveals important trade-offs between architectures that are not captured by clean-image benchmarks alone. These findings highlight the necessity of multi-level evaluation strategies combining overlap metrics, statistical testing, robustness analysis, and descriptor stability assessment for future benchmarking and clinically reliable deployment of nuclei segmentation systems. Full article

To achieve high-precision trajectory tracking control for an underwater hexapod robot, this paper proposes a hierarchical control architecture. Firstly, a multi-rigid-body dynamic model for the robot is established based on the Newton-Euler method and reasonably simplified. Secondly, a Central Pattern Generator (CPG) network with the Hopf oscillator as its core is designed to generate stable and coordinated crawling gaits. By introducing a steering parameter, a kinematic model connecting the CPG output is constructed. Furthermore, based on this dynamic and kinematic model, an upper-layer Model Predictive Controller (MPC) is designed. The optimized control quantities output by the MPC are mapped into the rhythmic parameters of the CPG network via a transfer function established by fitting experimental data, thus forming the complete MPC-CPG controller. Finally, the proposed method is validated through simulations of circular trajectory tracking. The results show that even in the presence of initial errors, the controller can converge rapidly, with trajectory position error consistently maintained within −0.1 m~0.1 m, and heading angle error confined to the range of −15~15°. The experiments fully demonstrate the effectiveness of the proposed MPC-CPG controller in ensuring trajectory tracking accuracy, motion smoothness, and system stability. Full article

This paper presents the design of a highly reconfigurable interdigital bandpass filter (BPF) developed through a three-stage design approach. In the first stage, the influence of four low-loss dielectric substrates on the filter response is systematically analyzed to identify the optimal configuration. The selected substrate demonstrates excellent performance, achieving an input return loss of −38 dB, an insertion loss of −0.9 dB at 4.30 GHz, and a wide passband corresponding to a bandwidth (BW) of 2.20 GHz. In the second stage, two variable capacitors were incorporated into the baseline geometry, enabling manual tuning of the center frequency ($f_0$) from 5.10 to 6.34 GHz, with $\left(S_{11}\right)$ better than −25 dB and ${(S}_{12})$ close to −0.60 dB. In the final stage, the capacitors were replaced by SMV1413 varactor diodes, transforming the design into a fully voltage-controlled tunable filter. This configuration provides continuous frequency agility from 4.70 to 5 GHz without modifying the physical structure, while achieving $\left(S_{11}\right)$ levels down to −40 dB and insertion loss as low as −0.7 dB. The proposed architecture offers a compact, low-loss, and electrically reconfigurable solution, making it a promising solution for next-generation RF front-ends, adaptive wireless systems, and cognitive radio applications. Two independent Electromagnetic solvers (EM) were employed to validate the filter’s performance: an EM based on the Finite Integration Technique and the Advanced Design System 2026 (ADS) solver using the Method of Moments (MoM). The close agreement between the results produced by both platforms confirms the accuracy and robustness of the proposed reconfigurable bandpass filter structure. Full article

Image-Based Overlay (IBO) equipment leverages optical reflection imaging principles, combined with focusing and alignment strategies to measure overlay marks. Among all measurement steps, the focal plane measurement of marks exerts the most critical impact on overlay accuracy, while the time consumed by focal plane detection directly determines the overall measurement throughput. To address the trade-off between accuracy and efficiency in advanced process nodes, this paper proposes an integrated optimization strategy encompassing optical hardware design and software algorithms. The hardware solution adopts a dual-wavelength, dual-detector architecture: optimal imaging wavelengths are selected independently for the previous-layer and current-layer marks, ensuring each layer achieves ideal imaging conditions without mutual interference. The software strategy employs a deep learning framework to simultaneously predict and adjust the horizontal position (alignment) and vertical defocus number of measured marks in real time with high precision, thereby securing the optimal imaging posture. By synergizing hardware-based optimal imaging conditions and software-based posture adjustment, this method effectively mitigates the impact of background noise and system aberrations, ultimately improving both the accuracy and efficiency of overlay measurement. Full article

Effective drug delivery in oncology is challenged by a hierarchy of biological barriers—from abnormal vasculature and dense stroma to cellular immunosuppression and specialized interfaces like the blood–brain barrier. This review provides a contemporary analysis of smart nanoformulations through the lens of a rational, stage-gated design pipeline. We first deconstruct the solid tumor microenvironment as a multi-tiered obstacle (systemic, stromal, cellular), establishing a barrier-specific foundation for nanocarrier design. The core of the review articulates an architectural toolkit, detailing how intrinsic nanoparticle properties precondition in vivo identity via the protein corona, which in turn informs the selection of advanced ligands for cellular targeting and programmed intracellular trafficking. This integrated framework sets the stage for exploring sophisticated applications, including endogenous and externally triggered responsive systems, bio-orthogonal activation, immuno-nanoformulations, and combination strategies aimed at overcoming multidrug resistance. By synthesizing these components into a cohesive design philosophy, this review moves beyond a catalog of advances to offer a blueprint for engineering next-generation nanotherapeutics. We critically assess the translational landscape and contend that this hierarchical design approach is essential for developing more effective, personalized, and clinically viable cancer treatments. Full article