Search Results (6,008)

Girth welds are susceptible to defects under high internal pressure and stress. While phased array ultrasonic testing (PAUT) is widely used for non-destructive evaluation, manual inspection remains inefficient and highly dependent on expertise. Furthermore, existing deep learning models often struggle with low accuracy and high complexity. This paper proposes a PAUT defect classification method based on YOLOv8. First, median filtering is employed for denoising, and the results show that noise is effectively reduced while preserving key features, achieving PSNR values of 35.132, 35.938, and 36.138 for slag inclusion, pores, and lack of fusion (LOF), respectively. Subsequently, the spatial enrichment algorithm (SEA) is applied to enhance image details without amplifying noise, yielding a PSNR of 33.71 and an SSIM of 0.96. Finally, the YOLOv8 model is implemented for defect recognition. Experimental results demonstrate that the proposed approach achieves a superior balance between precision and recall with high reliability. This method offers a robust and efficient solution for automated PAUT evaluation in practical engineering applications. Full article

Modern dentistry increasingly relies on light-curing units (LCUs) and lasers in essential clinical procedures such as composite resin polymerization, caries treatment, and periodontal therapy. This review aims to outline the evolution of light-emitting technologies and to assess their potential biological risks, with particular emphasis on effects on the visual system, oral tissues, and microbiome. The development of curing devices is presented chronologically, from the first-generation ultraviolet (UV-A) lamps introduced in the 1970s to current light-emitting diode (LED-LCU) systems and dental lasers (e.g., Er:YAG, Nd:YAG). The progressive increase in light intensity—now exceeding 3000 mW/cm²—has shortened curing times but simultaneously raised safety concerns. Major hazards include the so-called blue-light hazard, where exposure to high-energy visible (HEV) blue light may accelerate macular degeneration, and temperature elevations in the pulp chamber, which may damage the dentin–pulp complex. Laser radiation also exerts significant microbiological effects: Er:YAG and diode lasers demonstrate bactericidal activity against biofilms and oral pathogens (e.g., P. gingivalis), although therapeutic outcomes depend on wavelength, dose, and exposure time. Suboptimal parameters may lead to microbiome disturbances, whereas low-level laser therapy (LLLT; 600–1200 nm) supports tissue regeneration and helps restore microbial balance. The individualization of irradiation parameters, combined with thorough theoretical knowledge, operator expertise, and technical understanding of LCUs and lasers, is essential for maximizing clinical benefits while minimizing health risks and preserving oral microbiome homeostasis. Full article

Urban radio spectrum monitoring is becoming increasingly complex due to the rapid growth of wireless devices, unauthorized emissions, and dynamic electromagnetic environments in smart cities. Traditional spectrum analysis approaches, based on manual operation or static detection techniques, are no longer sufficient to ensure scalable, autonomous, and secure monitoring. The convergence of two emergent technologies—Large Language Models (LLMs) and the Model Context Protocol (MCP)—facilitates a fundamental shift in radio monitoring. We define this as the AICEBERG paradigm: a novel, stratified architecture where a high-level, intelligent agentic interface (the peak) abstracts the underlying complexity of SCPI-driven hardware integration and radio governance protocols (the foundational base). This autonomous framework provides the necessary objective rigor to audit the stochastic ‘ocean of electromagnetic waves’ characteristic of modern smart cities, ensuring a stable platform for regulatory enforcement amidst high-density signal interference. The proposed system implements a three-layer processing flow, enabling high-level natural language commands to be translated into validated and secure hardware actions on RF spectrum analyzers. A dual-server design separates operational execution from safety validation, ensuring controlled SCPI command handling, parameter verification, and instrument health monitoring. Experimental validation demonstrates the feasibility of autonomous measurement execution. The results show that the proposed architecture reduces human dependency, enhances reproducibility and lowers the expertise barrier required for RF spectrum surveillance. To the best of our knowledge, AICEBERG represents one of the first integrated frameworks to bridge LLMs with SCPI-compliant hardware through the MCP for autonomous radio governance. Full article

Background: Transcatheter aortic valve implantation (TAVI) is a viable alternative therapeutic approach for patients with severe aortic stenosis (AS), following technological innovations in transcatheter aortic valve systems and advances in clinical expertise, which aim to optimize valve hemodynamics. In this study, we aimed to compare early hemodynamic changes in different types of TAVI valves via two-dimensional echocardiography. Methods: This retrospective observational study examined patients with severe AS who underwent transfemoral TAVI. Patients were classified according to expansion mechanism (self-expanding valves (SEVs) or balloon-expandable valves (BEVs)) and leaflet position relative to the annulus (supra-annular valves (SAVs) or intra-annular valves (IAVs)). The implanted prostheses were Edwards SAPIEN XT valves (ESV, Edwards Lifesciences, Irvine, CA, USA), Medtronic valves (Core Valve-MCV and Evolut R, Medtronic, Minneapolis, MN, USA), Portico valves (St. Jude Medical, Saint Paul, MN, USA), and Myval valves (Meril Life Sciences, Vapi, India). Baseline two-dimensional transthoracic echocardiography (TTE) datasets were compared with post-TAVI measures obtained before discharge. Results: In total (n = 332), 275 (82.8%) patients were treated with SEVs, and 57 (17.2%) were treated with BEVs. In terms of leaflet position, 249 (75%) patients were treated with SAVs, and the remaining 83 (25%) patients were treated with IAVs. Transaortic gradients were comparable between patients treated with SEVs and BEVs. However, patients treated with IAVs exhibited significantly higher aortic maximum gradients (16 [13–21] mmHg vs. 14 [10–20] mmHg, p = 0.019) and mean gradients (9 [7–11] mmHg vs. 8 [5–10] mmHg, p = 0.014) compared to those receiving SAVs. Post-TAVI gradients were also compared based on each TAVI device. Although post-TAVI aortic maximum gradient was comparable among TAVI devices (p = 0.080), aortic mean gradient was significantly different among the valves (p = 0.006). Post hoc analyses demonstrated that the post-TAVI mean gradient was significantly lower in Medtronic CoreValve compared to the Myval (p = 0.013) and Portico (p = 0.030). No significant differences were observed in the frequency of perivalvular leak between the valve groups. Conclusions: We found that post-TAVI transaortic gradients of SEVs and BEVs were comparable; however, SAVs were associated with lower transaortic gradients than those of the IAVs. In addition, the frequency of ≥moderate PVL was comparable between the valve groups. Full article

Reproducible morphological profiling, particularly for drug discovery, has become an important tool for compound evaluation. Established workflows such as CellProfiler provide a widely adopted foundation for Cell Painting analysis. However, conventional pipelines often require substantial manual configuration and technical expertise, which can limit scalability and accessibility. In this study, a fully automated deep learning-based workflow is presented for segmentation-driven morphological profiling from raw microscopy data. Using a curated subset of the JUMP Cell Painting pilot dataset, ground-truth masks were generated and used to train a U-net–based segmentation model in the IKOSA platform. Post-processing strategies were introduced to improve instance separation and reduce segmentation artifacts. The final model achieved strong segmentation performance (precision/recall/AP up to 0.98/0.94/0.92 for nuclei), with an average runtime of 2.2 s per 1080 × 1080 image. Segmentation outputs enabled large-scale feature extraction, yielding 3664 morphological descriptors that showed high correlation with CellProfiler-derived measurements (normalized MAE: 0.0298). Feature prioritization further reduced redundancy to 1145 informative descriptors. These results demonstrate that automated deep learning pipelines can complement established Cell Painting workflows by reducing configuration overhead while maintaining compatibility with validated morphological profiling standards. The proposed workflow may help improve resource efficiency in drug discovery and personalized medicine. Full article

Lignocellulose represents an abundant repository of renewable carbon. Derived from various plant sources, it holds tremendous potential as a renewable and sustainable feedstock for the production of valuable chemicals and fuels. However, its solid fermentable compounds, cellulose and hemicellulose, are embedded within complex lignin structures and are therefore poorly accessible to microbial conversion. This paper describes an artificial rumen reactor (ARR) that uses anaerobic microbes from the cattle rumen to increase the release of fermentable carbon from recalcitrant biomass. We outline the development of an ARR for the efficient conversion of lignocellulosic grass into volatile fatty acids (VFAs), which are valuable precursors for the production of a range of bioproducts, including biofuels, biomaterials, and biochemicals. The ARR, a 4-L bioreactor equipped with a ceramic filtration unit, has been optimised and was operated for extended periods of continuous VFA production. Across distinct short- and long-term observation periods, and independent of the cow from which the rumen microbes originated, the bioreactor demonstrated the ability to sustain VFA production, indicating robustness and stability. At an input of 60–80 g dry grass d⁻¹, the system produced approximately 6 mol VFA per kg of dry matter input (DMI). The decoupling of the Solid Retention Time (SRT; 10 days) and the Liquid Retention Time (LRT; 0.5 days) prevented inhibition of the VFA production. The VFA profile was dominated by acetic and propionic acids, comprising 68% and 19%, respectively, with butyric acid and minor VFAs accounting for the remainder. The application of low oxygen levels (<10%) in the reactor via limited aeration did not affect the VFA yield or its profile. Full article

Objective: We aimed to determine the ability of prenatal ultrasound to detect velamentous cord insertion (VCI) in twin pregnancies and identify factors influencing diagnostic sensitivity. Methods: This single-center retrospective study included twins delivered between April 2018 and March 2024. We excluded monochorionic monoamniotic twins, those without chorionicity or umbilical cord insertion data, and fetuses that died in utero. Umbilical cord insertion sites assessed by second-trimester transabdominal ultrasound (16 + 0 to 21 + 6 weeks of gestation) using color Doppler imaging were classified as normal, marginal, or velamentous. The results of postnatal macroscopic examinations served as reference standards. We calculated accuracy, sensitivity, specificity, positive (PPV) and negative (NPV) predictive values. The effects of examiner expertise, chorionicity, placental location, ultrasound device, and maternal body mass index (BMI) on diagnostic sensitivity were analyzed in subgroups. Results: We confirmed VCI in 45 (8.8%) of 514 delivered fetuses. Prenatal ultrasound correctly identified 14 VCI cases. Sensitivity, specificity, PPV, and NPV were 31.1% (14/45), 98.9% (464/469), 73.7% (14/19), and 93.7% (464/495), respectively. The overall accuracy was 93.0% (478/514). Sensitivity was significantly higher when ultrasound specialists conducted examinations compared with non-specialists and when twins were monochorionic diamniotic twins than dichorionic. Anterior placental location and high-performance ultrasound equipment were also associated with increased sensitivity, but were not statistically significant. Maternal BMI did not affect diagnostic sensitivity. Conclusions: Prenatal ultrasonographic detection of VCI in twin pregnancies has high specificity but limited sensitivity. Diagnostic performance was influenced by examiners’ experience and chorionicity. Routine assessment of cord insertion sites and targeted training might improve detection and support the optimized perinatal management of twin pregnancies. Full article

This paper proposes a human-centric digital twin (DT) framework balancing energy efficiency with occupant well-being in existing buildings, addressing the lack of actionable insights in data-driven facility management and comfort issues common in fully automated systems. A “Human-in-the-loop” approach using dual-KPIs integrates real-time IoT data and visualization to evaluate sustainable energy use via Indoor Environmental Quality (IEQ). A novel occupancy-inference method tracks efficiency in legacy buildings without granular metering, implemented through a case study of 26 office rooms. Results indicate that the framework successfully identifies significant energy wastage and comfort anomalies without compromising well-being. Integrating real-time analytics with human oversight enables more resilient management than fully automated alternatives, particularly for detecting non-operational heating waste. The occupancy inference method was validated against ground truth, achieving 81% accuracy, with limitations regarding decay lag discussed. This research offers a cost-effective diagnostic tool for legacy buildings lacking sub-metering, lowering DT adoption barriers, and shifting maintenance from reactive to data-driven strategies. The framework leverages human expertise and infers occupancy-normalized energy metrics from standard IEQ sensors, proposing a human-centric DT framework to bridge the gap between raw sensor data and actionable facility management insights. Full article

Background: Accurate differentiation of kidney diseases such as cysts, tumors, stones, and normal tissue from computed tomography (CT) images remains challenging due to overlapping visual characteristics and variability in data distributions. While deep learning approaches have shown promising results, many existing studies rely on image-level data splitting and focus primarily on accuracy, which may lead to overly optimistic performance and limited clinical reliability. Methods: This study proposes KDH-Net (Kidney Disease Hybrid Network), a hybrid deep learning framework for multiclass kidney disease characterization that integrates EfficientNetB0, ResNet50, and MobileNetV2 through feature-level fusion. A two-stage training strategy is adopted to enhance optimization stability. To ensure realistic performance assessment, experiments on the primary dataset are conducted under a patient-level evaluation protocol, eliminating potential data leakage. The framework further incorporates calibration analysis, statistical validation, and explainable artificial intelligence to evaluate prediction reliability and interpretability. Results: On the patient-level dataset, KDH-Net achieves an overall accuracy of 0.93 with a macro-average F1-score of 0.91, demonstrating balanced performance across all classes. Confidence analysis indicates meaningful alignment between prediction confidence and correctness, while Grad-CAM visualizations highlight anatomically relevant regions associated with each class. Conclusions: The results demonstrate that KDH-Net provides a stable, reliable, and interpretable framework for kidney CT characterization. The proposed system is designed to support clinical decision-making by offering trustworthy predictions under realistic evaluation conditions, rather than replacing clinical expertise. Full article

Background/Objectives: Cardiovascular diseases are the leading cause of death worldwide. Therefore, early diagnosis and treatment of these diseases are of critical importance. Stethoscopes are the easiest and fastest medical devices for the initial diagnosis of cardiovascular diseases. However, interpreting heart sounds requires considerable expertise. The use of artificial intelligence in healthcare for decision support has increased and become popular recently. Methods: The popular 2016 PhysioNet/CinC Challenge dataset, consisting of phonocardiogram (PCG) signals, was used to implement the proposed approach. Spectrogram and continuous wavelet transform (CWT) images of the PCG signals were first generated. This increased the distinguishability of the data in terms of both time and frequency components. These two-input images were tested on the developed Dual Time–Frequency Swin Transformer–ConvNeXt Attention Network (DTF-STCANet) model. To further improve classification accuracy, the Weighted KNN algorithm was preferred during the classification phase. Results: With the proposed approach, a 99.29% classification accuracy was achieved. Performance was compared with other state-of-the-art models. Conclusions: The proposed approach, through the integration of PCG signals with artificial intelligence, further strengthens the concept of early diagnosis of heart disease. Full article

Background and Objectives: Neoadjuvant chemoradiotherapy is a key component of the treatment strategy for locally advanced rectal cancer (LARC), both through its direct impact on oncological prognosis and by increasing the likelihood of sphincter-preserving surgery. Oncological prognosis improves dramatically following a complete pathological response to neoadjuvant therapy. Identifying predictors of response to neoadjuvant therapy has been a challenge over the past two decades, and these factors have not been fully identified. This study aimed to analyze the clinical, biological, and therapeutic factors associated with tumor response following neoadjuvant therapy in patients with locally advanced rectal cancer, with the aim of identifying independent predictors of the absence of a complete pathological response and optimizing personalized treatment strategies. Materials and Methods: This retrospective study included a cohort of 122 patients (81 men and 41 women), with a mean age of 63.5 years, diagnosed with locally advanced rectal cancer at two centers with expertise in colorectal surgery between January 2018 and December 2023. Patients received neoadjuvant treatment in two regimens: long-course chemoradiotherapy with oral radiosensitizing chemotherapy (82 patients) and total neoadjuvant therapy consisting of chemoradiotherapy followed by consolidation chemotherapy (40 patients). A series of clinical, biological, and therapeutic variables was analyzed for their association with pathological responses. Results: According to the Ryan score, the overall complete response rate following neoadjuvant therapy was 17.2%. pCR was observed more frequently in patients treated with total neoadjuvant therapy than in those treated with standard chemoradiotherapy. Elevated pre-treatment CEA levels were independently associated with a higher risk of unfavorable tumor response. The radiation dose and interval between completion of radiotherapy and surgery were significantly associated with tumor regression. Conclusions: These results underscore the importance of personalizing neoadjuvant therapy to improve cancer prognosis. Furthermore, optimizing tumor regression could lead to the potential expansion of sphincter-preserving resection techniques, which would have a direct and significant impact on the quality of life of these patients. Full article

Multi-block structured mesh is widely used for high-precision aerodynamic simulation, but mesh blocking usually requires substantial manual intervention, which is time-consuming and demands a high level of user expertise. In this study, a potential field-based blocking algorithm for mesh generation around an aircraft is proposed, and a corresponding multi-block grid generation workflow is established. First, the hyperbolic partial differential equation (PDE) method is used to march boundary layer grids from the body surface. Next, the potential field is solved on an unstructured background grid, and the grid topology is flexibly designed by adjusting boundary conditions. The gradient lines of the potential field are then determined and employed to partition the external domain into blocks. Finally, the elliptic PDE method is applied to generate structured grids within each sub-block. A low-aspect-ratio flying-wing configuration is adopted as the test case. Structured grids of both H-type and O-type topologies are generated and compared with the benchmark grid released by the China Aerodynamics Research and Development Center (CARDC). The grid quality analysis and aerodynamic calculation results demonstrate that the two generated grids possess good quality, and the computational results show satisfactory agreement with experimental data. The O-type mesh yields more accurate predictions for the lift coefficient and pitching moment coefficients. Furthermore, two test cases, namely a rocket sled and a V-tail aircraft, are presented to demonstrate that the proposed method can flexibly design either O-type or H-type topologies to accommodate different geometric characteristics. In summary, the proposed method enables efficient generation of high-quality multi-block structured grids for the configurations examined in this study. Full article

Agriculture in the Pacific is driven primarily by small-scale private farmers, many of whom do not have access to soil testing services or advice, nor the means to interpret analytical results into soil management and agronomic recommendations. Soil degradation through the process of acidification poses a significant risk to food and income security as it directly threatens crop productivity. The nutritional quality of food crops may also be affected through sub-optimal nutrient uptake by plants and nutrient imbalances. The dataset reported here provides a useful platform for the development of a decision-support tool (DST) that will assist Fiji farmers in understanding and managing soil pH and soil acidity. The DST will enable making informed decisions about liming to help correct soil pH. To support this development, historical soil pH data available from the Pacific Soils Portal were combined with updated analyses of agricultural soils from 17 locations in Viti Levu Island (Fiji) collected during a field campaign undertaken in August 2025. The soils were sampled at two depth intervals (0–15 and 15–30 cm) and analyzed for pH using a variety of methods. These methods included direct field measurements using a portable pH-meter as well as traditional laboratory determinations. Of the soils sampled, it was found that most soils exhibited pH levels below 7, which were observed for both depth intervals. Across all samples taken in 2025, it was found that 54.3% of them had soil pH < 5, 38.6% had soil pH between 5 and 6, and 7.1% had pH > 6 (based on soil pH_1:5 soil-to-water method). Depending upon specific land uses, climate and cropping intensity, it was recommended that routine liming be built into soil fertility management programs to help farmers overcome soil acidity-related constraints to production. Liming frequency, timing of application and application rate will need to be determined for specific soil and cropping situations; however, it was suggested that soil pH was not changed by more than 1 unit each time lime was applied. Such an approach should reduce the risk of soil organic matter loss through accelerated mineralization, which would be challenging to restore in that environment if soils remained under continuous cropping. The analytical information contained in this article expanded and updated the datasets available in the Pacific Soils Portal. Furthermore, this work provided an opportunity to build analytical expertise in aspects of soil chemistry at local organizations to support academic and extension activities as well as the ongoing development of the Pacific Soils Portal. Full article

Building Information Modelling (BIM) has emerged as a transformative approach for improving efficiency, coordination, and sustainability in the construction industry; however, its adoption in developing regions remains limited. In the Kurdistan Region of Iraq (KRG), BIM implementation—particularly within the residential construction sector—remains at an early stage and lacks comprehensive empirical investigation. This study aims to assess stakeholders’ perspectives on current BIM practices, identify key adoption barriers, and develop a context-specific strategic framework to support BIM implementation. A mixed-method research design was employed, incorporating literature review, expert validation through semi-structured interviews, and a structured questionnaire survey. A total of 319 valid responses were analyzed using descriptive statistics, Relative Importance Index (RII), Cronbach’s alpha for reliability, Spearman’s rank correlation, independent samples t-tests, and one-way ANOVA. In addition to ranking barriers, an inter-barrier correlation analysis was conducted to examine the relationships, clustering patterns, and hierarchical structure of BIM adoption challenges. The results indicate that while BIM awareness is moderately established among stakeholders, its practical application remains limited, particularly beyond the design phase. The most critical barriers include lack of training and expertise, absence of regulatory frameworks and standards, insufficient government support, and financial constraints. The correlation analysis reveals that these barriers are interdependent, with policy and institutional deficiencies acting as root drivers influencing technical, financial, and awareness-related challenges. Based on these findings, the study proposes a four pillar conceptual strategic framework encompassing human capital development, regulatory and standardization enablement, awareness and demand generation, and organizational and collaborative enhancement. The framework is explicitly derived from empirical results, linking barrier clusters to prioritized strategies, thereby enhancing its practical applicability. This study contributes to the existing literature by providing one of the first multi-province empirical assessments of BIM adoption in the KRG residential sector, integrating statistical validation with strategic development, and offering transferable insights for other developing regions at a similar stage of BIM adoption. Full article

Sustainable pedagogical reform requires more than teacher preparedness; it depends on how school leaders interpret and coordinate the conditions that enable change. This focus is particularly critical in contexts where Problem-Based Learning (PBL) is introduced within predominantly traditional, exam-oriented pedagogical environments, requiring careful consideration of leadership’s perception of system-level readiness to support such shifts. This study investigates how Chinese K–12 school leaders conceptualize readiness for institution-wide implementation of PBL. Using Q methodology with 42 school leaders, four distinct leadership logics were identified: leadership-mediated cultural readiness through recognition, belief-driven pedagogical practice, externally anchored system-level readiness, and experientially grounded cultural readiness. These viewpoints reveal different ways leaders prioritize cultural alignment, belief formation, structural coordination, and experiential learning when organizing reform conditions. Despite these differences, participants showed several areas of shared positioning, particularly around coordination, expertise-based responsibility distribution, evaluation alignment, and adaptive responses to reform conditions. The findings extend change readiness research beyond teacher-focused perspectives by demonstrating how leaders interpret readiness as a multidimensional and system-level phenomenon. By illuminating distinct leadership logics for coordinating reform within centralized governance contexts, this study highlights the importance of aligning beliefs, professional relationships, institutional structures, and student learning improvement goals to support sustainable pedagogical transformation. Full article