Search Results (13,560)

Background: Cancer poses a significant health risk to humans, with prostate cancer (PCa) being the second most common and deadly form among men, following lung cancer. Each year, it affects over a million individuals and presents substantial diagnostic challenges due to variations in tissue appearance and imaging quality. In recent decades, various techniques utilizing Magnetic Resonance Imaging (MRI) have been developed for identifying and classifying PCa. Accurate classification in MRI typically requires the integration of complementary feature types, such as deep semantic representations from Convolutional Neural Networks (CNNs) and handcrafted descriptors like Histogram of Oriented Gradients (HOG). Therefore, a more robust and discriminative feature integration strategy is crucial for enhancing computer-aided diagnosis performance. Objectives: This study aims to develop a multi-stage hybrid learning model that combines deep and handcrafted features, investigates various feature reduction and classification techniques, and improves diagnostic accuracy for prostate cancer using magnetic resonance imaging. Methods: The proposed framework integrates deep features extracted from convolutional architectures with handcrafted texture descriptors to capture both semantic and structural information. Multiple dimensionality reduction methods, including singular value decomposition (SVD), were evaluated to optimize the fused feature space. Several machine learning (ML) classifiers were benchmarked to identify the most effective diagnostic configuration. The overall framework was validated using k-fold cross-validation to ensure reliability and minimize evaluation bias. Results: Experimental results on the Transverse Plane Prostate (TPP) dataset for binary classification tasks showed that the hybrid model significantly outperformed individual deep or handcrafted approaches, achieving superior accuracy of 99.74%, specificity of 99.87%, precision of 99.87%, sensitivity of 99.61%, and F1-score of 99.74%. Conclusions: By combining complementary feature extraction, dimensionality reduction, and optimized classification, the proposed model offers a reliable and generalizable solution for prostate cancer diagnosis and demonstrates strong potential for integration into intelligent clinical decision-support systems. Full article

In urban underground space construction using shield tunnelling, the geological conditions ahead of the tunnel face are often uncertain. Without timely and accurate classification of the ground type, mismatches in operational parameters, uncontrolled costs, and schedule risks are likely to occur. Using observations from an earth pressure balance (EPB) project on an urban railway, a data-driven classification framework is developed that integrates shield tunnelling operating measurements with physically derived quantities to discriminate among soft soil, hard rock, and mixed strata. Principal component analysis (PCA) is performed on the training set, followed by a systematic comparison of tree-based classifiers and hyperparameter optimization strategies to explore the attainable performance. Under unified evaluation criteria, a categorical bosting (CatBoost) model optimized by a Nevergrad combination strategy (NGOpt) attains the highest test accuracy of 0.9625, with macro-averaged precision and macro-averaged recall of 0.9715 and 0.9716, respectively. To mitigate optimism from single-point estimates, stratified bootstrap intervals are reported for the test set. A Monte Carlo experiment applies independent perturbations to the PCA-transformed features, producing low label-flip rates across the three classes, with only minor changes in probability calibration metrics, which suggests consistent decisions under sensor noise and sampling bias. Overall, within the scope of the considered EPB project, the study delivers a compact workflow that demonstrates the feasibility of uncertainty-aware ground-type classification and provides a methodological reference for developing decision-support tools in underground tunnel construction. Full article

Produced from the parenchymatous tissue of the stem pith of Tetrapanax papyrifer, the material known as pith paper served as a distinctive support medium for Chinese export paintings during the 19th and early 20th centuries. Today, it is commonly found in collections worldwide. Due to its inherently fragile structure, conservation interventions are often necessary. However, the material’s chemical composition and deterioration mechanisms remain poorly understood, which not only complicates treatment decisions but also undermines preventive conservation efforts. This study presents a systematic investigation into the anatomical structure and ageing behaviour of pith paper using a multi-analytical approach. Optical and scanning electron microscopy revealed a preserved honeycomb-like cellular architecture composed of thin-walled, entirely of non-lignified parenchyma cells, which contributes to the material’s mechanical fragility. Artificial ageing experiments showed a significant loss of flexibility, increased yellowing, and a decline in pH with ageing time. Infrared spectroscopy identified molecular changes consistent with cellulose chain scission, with decreases in O–H and C–O–C absorptions revealing acid-hydrolysis-driven breakdown, while colourimetry pointed to the formation of chromophoric degradation products. These findings offer a foundational understanding of pith paper’s vulnerabilities and provide essential insights for the development of informed conservation and storage strategies. Full article

The digital transformation of manufacturing is accelerating through the integration of artificial intelligence (AI), particularly via real-time predictive models. These models enable manufacturers to transition from reactive to proactive strategies, intelligent optimization and decision-making. Within the frameworks of Industry 4.0 and Industry 5.0, which emphasize technologies such as cyber-physical systems, cloud computing, and human-centric innovation, AI-driven data models are pivotal for achieving smart, adaptive, and sustainable production systems. This paper investigates the impact of AI-based predictive modeling on manufacturing digitalization and its future potential. It examines how these models contribute to advanced frameworks such as online process advisory systems, digital shadows, and digital twins, while addressing their limitations and implementation challenges. Furthermore, the study reviews current practices in real-time data modeling across manufacturing processes—including direct-chill casting—supported by real-world case studies. These examples illustrate both the practical benefits and technical hurdles of deploying AI in dynamic industrial environments. Full article

Background: Cervical cancer remains a major global health challenge, ranking fourth among malignancies in women, with an estimated 660,000 new cases and 350,000 deaths in 2022. Despite advances in vaccination and screening, incidence and mortality remain disproportionately high in low- and middle-income countries. The disease is strongly linked to persistent infection with high-risk human papillomavirus (HPV) types, predominantly HPV 16 and 18, whose E6 and E7 oncoproteins drive cervical intraepithelial neoplasia (CIN) and invasive cancer. This review summarizes current evidence on clinically relevant biomarkers in HPV-associated CIN and cervical cancer, emphasizing their role in screening, risk stratification, and disease management. Methods: We analyzed the recent literature focusing on validated and emerging biomarkers with potential clinical applications in HPV-related cervical disease. Results: Biomarkers are essential tools for improving early detection, assessment of progression risk, and personalized management. Established markers such as p16 immunostaining, p16/Ki-67 dual staining, and HPV E6/E7 mRNA assays increase diagnostic accuracy and reduce overtreatment. Prognostic indicators, including squamous cell carcinoma antigen (SCC-Ag) and telomerase activity, provide information on tumor burden and recurrence risk. Novel approaches—such as DNA methylation panels, HPV viral load quantification, ncRNAs, and cervico-vaginal microbiota profiling—show promise in refining risk assessment and supporting non-invasive follow-up strategies. Conclusions: The integration of validated biomarkers into clinical practice facilitates more effective triage, individualized treatment decisions, and optimal use of healthcare resources. Emerging biomarkers, once validated, could further improve precision in predicting lesion outcomes, ultimately reducing the global burden of cervical cancer and improving survival. Full article

To address the issues of efficiency and robustness in UAV trajectory planning under complex environments, this paper proposes a Graph Multi-Head Attention Policy Optimization Network (G-MAPONet) algorithm that integrates Graph Attention (GAT), Multi-Head Attention (MHA), and Group Relative Policy Optimization (GRPO). The algorithm adopts a three-layer architecture of “GAT layer for local feature perception–MHA for global semantic reasoning–GRPO for policy optimization”, comprehensively achieving the goals of dynamic graph convolution quantization and global adaptive parallel decoupled dynamic strategy adjustment. Comparative experiments in multi-dimensional spatial environments demonstrate that the Gat_Mha combined mechanism exhibits significant superiority compared to single attention mechanisms, which verifies the efficient representation capability of the dual-layer hybrid attention mechanism in capturing environmental features. Additionally, ablation experiments integrating Gat, Mha, and GRPO algorithms confirm that the dual-layer fusion mechanism of Gat and Mha yields better improvement effects. Finally, comparisons with traditional reinforcement learning algorithms across multiple performance metrics show that the G-MAPONet algorithm reduces the number of convergence episodes (NCE) by an average of more than 19.14%, increases the average reward (AR) by over 16.20%, and successfully completes all dynamic path planning (PPTC) tasks; meanwhile, the algorithm’s reward values and obstacle avoidance success rate are significantly higher than those of other algorithms. Compared with the baseline APF algorithm, its reward value is improved by 8.66%, and the obstacle avoidance repetition rate is also enhanced, which further verifies the effectiveness of the improved G-MAPONet algorithm. In summary, through the dual-layer complementary mode of GAT and MHA, the G-MAPONet algorithm overcomes the bottlenecks of traditional dynamic environment modeling and multi-scale optimization, enhances the decision-making capability of UAVs in unstructured environments, and provides a new technical solution for trajectory planning in intelligent logistics and distribution. Full article

Approximately 80% of invasive breast cancers are classified as human epidermal growth factor receptor 2 (HER2)-negative; however, many of these tumors have detectable levels of HER2 surface expression. Trastuzumab deruxtecan (T-DXd) is a HER2-directed antibody-drug conjugate with a membrane-permeable payload that is cytotoxic to both HER2-expressing tumor cells and neighboring cells via the bystander antitumor effect. T-DXd has shown significant antitumor activity in clinical trials for patients with HER2-positive (immunohistochemistry [IHC] 3+ or IHC 2+/in situ hybridization [ISH]+) breast cancer. In addition, the results of the DESTINY-Breast04 trial demonstrated the clinical benefit of T-DXd in patients with HER2-low (IHC 1+ or IHC 2+/ISH−) breast cancer after receiving prior chemotherapy. DESTINY-Breast06 demonstrated the clinical benefit of T-DXd in patients with hormone receptor (HR)-positive, HER2-low (IHC 1+ or IHC 2+/ISH−), and HER2-ultralow (IHC 0 with membrane staining) metastatic breast cancer who had not received prior chemotherapy in the advanced setting. These results validate the need for a standard-of-care diagnostic test to identify HER2-low and HER2-ultralow expression levels in patients with metastatic breast cancer to guide therapeutic decision-making. Furthermore, effective treatment sequencing strategies and adverse event management are essential for maximizing patient benefit. This review presents the identification of HER2-low and HER2-ultralow breast cancer, sequencing of T-DXd with other treatments, and management of common or clinically significant adverse events reported with T-DXd. Full article

In open wireless communication channels, the combined effects of random pulse jamming and multipath-induced time-varying fading significantly degrade the reliability and efficiency of information transmission. Particularly in highly dynamic scenarios such as unmanned aerial vehicle (UAV) communications, existing Q-learning-based anti-jamming methods often rely on idealized channel assumptions, leading to mismatched “transmit/silence” decisions under fading conditions. To address this issue, this paper proposes a Q-learning and time-varying fading channel-aware anti-jamming method against random pulse jamming. In the proposed framework, a fading channel model is incorporated into Q-learning, where the state space jointly represents timeslot position, jamming history, and channel sensing results. Furthermore, a reward function is designed by jointly considering jamming power and channel quality, enabling dynamic strategy adaptation under rapidly varying channels. A moving average process is applied to smooth simulation fluctuations. The results demonstrate that the proposed method effectively suppresses jamming collisions, enhances the successful transmission rate, and improves communication robustness in fast-fading environments, showing strong potential for deployment in practical open-channel applications. Full article

Background: Muscle-invasive bladder cancer (MIBC) represents a highly aggressive malignancy associated with significant morbidity and mortality. The current standard treatment, which includes radical cystectomy and platinum-based chemotherapy, is burdened by high toxicity and a substantial risk of relapse. For this reason, over the past decade, novel therapeutic strategies involving immune checkpoint inhibitors (ICIs), antibody–drug conjugates (ADCs), and targeted therapies have been investigated. This review aims to summarize current clinical evidence and ongoing trials evaluating these approaches in the perioperative setting. Methods: A systematic search was conducted using PubMed, EMBASE, and Cochrane databases, along with abstracts from major oncology conferences (ASCO, ESMO, SGO). Clinical trials assessing ICIs, ADCs, and targeted therapies, either alone or in combination with each other or with chemotherapy, in MIBC, were included. Results: Several early-phase and phase III trials have investigated the perioperative management of MIBC. Various studies evaluated the addition of ICIs to standard chemotherapy, demonstrating promising results in terms of pathological complete response. In parallel, the encouraging outcomes with ICIs and ADCs alone in the neoadjuvant or adjuvant setting paved the way for their combination in integrated strategies. Biomarker-driven approaches, based on circulating tumor DNA and specific genomic alterations, are being actively explored to improve patient selection and personalize treatment. Conclusions: ICIs, ADCs, and targeted therapies are reshaping the therapeutic landscape of MIBC. While early results are promising, further data and biomarker validation are essential to establish their definitive role and guide clinical decision-making in the perioperative setting. Full article

The rapid growth of electric vehicles has intensified the spatial mismatch between the layout of charging infrastructure and user demand, resulting in a structural contradiction in which “local oversupply” and “local shortages” coexist. To systematically diagnose and optimize this issue, this study develops an innovative analytical framework for a “residential area–charging infrastructure” collaborative service network and conducts an empirical analysis using Hongshan District in Wuhan as a case study. The framework integrates actual facility utilization data, complex network analysis, and spatial clustering methods. The findings reveal that the collaborative service network in the study area is overall sparse, exhibiting a distinct “core–periphery” structure, with noticeable patterns of resource concentration and isolation. Residential areas can be categorized into three types based on their supply–demand characteristics: efficient-collaborative, transitional-mixed, and low-demand peripheral areas. The predominance of the transitional-mixed type indicates that most areas are currently in an unstable state of supply–demand adjustment. A key systemic mechanism identified in this study is the significant “collaborative reinforcement effect” between facility utilization rates and network centrality. Building on these insights, we propose a hierarchical optimization strategy consisting of “overall network optimization—local cluster coordination—individual facility enhancement.” This ultimately forms a comprehensive decision-support framework for “assessment—diagnosis—optimization,” providing scientific evidence and new solutions for the precise planning and efficient operation of urban charging infrastructure. Full article

To address the issues of high energy consumption and high carbon emissions associated with the steam injection development of ultra-heavy oil in China, technological exploration focusing on electrical heating and solvent substitution was conducted. Firstly, experiments on the heat transfer and temperature rise characteristics in the near-wellbore formation via electrical heating revealed its feasibility. Considering that ultra-heavy oil reservoirs in China suitable for Steam-Assisted Gravity Drainage (SAGD) have already been converted to SAGD production, and considering the certain safety risks of solvent extraction, a development strategy of SAGD—Electrical Heating Solvent Extraction—SAGD was formulated. A multi-stage drainage theoretical model coupling SAGD with electrical heating solvent extraction was established. The similarity criteria for 3D-scaled physical simulation of electrical-heating-assisted production were derived. Through three-stage (SAGD—Electrical Heating Solvent Extraction—SAGD) scaled physical simulation experiments, the development performance of converting a SAGD-developed reservoir to thermal solvent extraction was analyzed. Results indicate that the higher the oil content in the electrically heated wellbore and nearby formation, the faster the heat transfer rate. This confirmed the decision to conduct experiments on electrical-heating-assisted solvent extraction (without steam injection) in SAGD-developed reservoirs. After the SAGD steam chamber reaches the top, switching to electrical heating solvent extraction results in a drainage zone along the flanks of the horizontal section comprising: a high-temperature zone of vaporized solvent from electrical heating, a medium-low temperature oil dissolution zone from the solvent, and an untouched zone. Along the horizontal section, it is divided into a solvent chamber rising zone, a slow expansion zone, and a rapid expansion zone. Experiments confirmed that electrical heating can vaporize the solvent, continuously expanding the drainage chamber scale. Furthermore, the solvent continues to function in the subsequent SAGD stage, increasing the recovery factor from 64.4% to 71.2%, an improvement of 6.9%. The established multi-stage coupled drainage theoretical model, compared with experimental and analytical calculations, showed an overall agreement rate of 95.3%, and can be used for production prediction in electrical-heating-assisted solvent extraction composite recovery. Full article

This study examines how consumers’ perceptions of threat severity and anxiety during public health crises influence their motivation to protect themselves and, subsequently, their intentions to purchase protection insurance products. Drawing on Protection Motivation Theory (PMT), we develop an integrated framework that links cognitive risk assessments and emotional responses to financial protection decisions. Using survey data collected from 437 respondents in Taiwan during the COVID-19 pandemic, the research model is tested through partial least squares structural equation modeling (PLS-SEM). The empirical results indicate that both perceived severity and anxiety significantly enhance protection motivation, with perceived severity exerting a stronger effect. These two antecedents also directly strengthen consumers’ intentions to purchase protection insurance. Furthermore, protection motivation partially mediates the effects of perceived severity and anxiety on purchase intention. These findings extend the application of PMT to the financial and insurance domains by demonstrating how cognitive and affective factors jointly shape demand for protection insurance in high-risk environments. The practical implications of these results for insurers include risk communication strategies, product positioning, and the development of crisis-responsive insurance solutions. Full article

Background: Due to the complexity of venous ulcer treatment, the role of nurses is critical, and artificial intelligence, particularly artificial neural networks of the Multilayer Perceptron type, can be effective tools that support professionals with objective, real-time evaluation. Thus, the present study aims to develop a network to assist in nurse decision-making regarding topical therapies for the treatment of venous ulcers. Methods: A methodological study with a technological focus and quantitative approach was conducted. The Unified Process methodology model was used, based on the Rational Unified Process strategy, following four phases: conception, elaboration, construction, and transition. Results: The development of the artificial neural network involved the collaboration of three specialists who evaluated clinical cases and images of venous ulcers to identify the topical therapies used in their clinical practice. A total of 23 dressings were selected, studied, and grouped into evaluation protocols to create the neural network flowchart, which defined the structure of the network. This network was then used by 13 nurses through the VenoTEC app (version 1.2, developed by the authors, Natal, Brazil). Conclusions: The software developed showed promising results in the initial evaluations conducted. The network achieved the highest accuracy in the initial tests and received a very good usability rating from the nurses who participated in the evaluation. The small dataset limits the generalization capability of the findings. Further studies are needed with additional datasets and populations. Full article

Big data and artificial intelligence technologies are driving a paradigm shift in smart farming, yet intelligent decision-making faces critical bottlenecks. At the data level, challenges include fragmentation, high acquisition costs, and inadequate secure sharing; at the model level, issues involve regional heterogeneity, weak adaptability, and insufficient explainability. To address these, this paper systematically reviews global research to establish a theoretical framework spanning the entire production cycle. Regarding data governance, trends favor federated systems with unified metadata and layered storage, utilizing technologies like federated learning for secure lifecycle management. For decision-making, approaches are evolving from experience-based to data-driven intelligence. Pre-harvest planning now integrates mechanistic models and transfer learning for suitability and variety optimization. In-season management leverages deep reinforcement learning (DRL) and model predictive control (MPC) for precise regulation of seedlings, water, fertilizer, and pests. Post-harvest evaluation strategies utilize spatio-temporal deep learning architectures (e.g., Transformers or LSTMs) and intelligent optimization algorithms for yield prediction and machinery scheduling. Finally, a staged development pathway is proposed: prioritizing standardized data governance and foundation models in the short term; advancing federated learning and human–machine collaboration in the mid-term; and achieving real-time, ethical edge AI in the long term. This framework supports the transition toward precise, transparent, and sustainable smart agriculture. Full article

Regional disparities pose persistent challenges for balanced and sustainable development in Türkiye, where provinces exhibit prominently heterogeneous socioeconomic structures, capacities, and investment needs. This study proposes an integrated, data-driven framework for allocating public investments across provinces by jointly addressing development efficiency and spatial equity. A dataset of 109 indicators for 81 provinces was compiled and standardized, and Principal Component Analysis, followed by multiple clustering algorithms (K-Means, Gaussian Mixture Model, Fuzzy C-Means), was used to derive robust provincial development profiles. National policy priorities were quantified through a document-based assessment of the 12th Development Plan (2024–2028), enabling the construction of nine strategic investment categories aligned with national objectives. These components were incorporated into a multi-objective optimization model formulated using the ε-constraint method, where total utility is maximized subject to an adjustable equity constraint based on a Gini-like parameter. Results reveal a clear efficiency–equity trade-off: low inequality tolerance yields uniform but low-return allocations, whereas relaxed equity constraints amplify concentration in high-capacity metropolitan provinces. Intermediate equity levels (G = 0.3–0.5) generate the most balanced outcomes, supporting both development potential and spatial cohesion. The proposed framework offers a transparent, reproducible decision support tool for more equitable and strategy-aligned public investment planning in Türkiye. Full article