Search Results (1,502)

The terahertz (THz) frequency range holds critical importance for next-generation, wireless communications and biomedical sensing applications. However, conventional metamaterial design approaches suffer from computationally intensive simulations and optimization processes that can extend over several months. This work presents an intelligent inverse design framework integrating deep neural network (DNN) surrogate modeling with success-history-based adaptive differential evolution (SHADE) for tunable graphene-based THz metasurfaces. Our DNN surrogate model achieves an exceptional coefficient of determination (R² = 0.9984) while providing a four-order-of-magnitude acceleration compared with conventional electromagnetic solvers. The SHADE-integrated framework demonstrates 96.7% accuracy in inverse design tasks with an average convergence time of 10.2 s. The optimized configurations exhibit significant tunability through graphene Fermi level modulation, as validated by comprehensive electromagnetic field analysis. This framework represents a significant advancement in automated electromagnetic design and establishes a robust foundation for intelligent photonic systems across diverse frequency regimes. Full article

To address the challenges in multimodal information integration and the inefficiency of knowledge transfer in the construction technology disclosure of deep foundation pit projects, an intelligent generation method based on graph rule reasoning and template mapping was proposed. First, a multi-level domain knowledge structure model was constructed by designing domain concepts and relationship types using the Work Breakdown Structure (WBS). Second, entity and attribute extraction was performed using regular expressions and the BERT-BiLSTM-CRF model, while relationship extraction was conducted based on text structure combined with the BERT-CNN model. For image and video data, cross-modal data chains were built by adding keyword tags and generating URLs, utilizing semantic association rules to form a multimodal knowledge graph of the domain. Finally, based on graph reasoning and template mapping technology, the intelligent generation of construction disclosure schemes was realized. The case verification results showed that the proposed method significantly improved the structural integrity, procedural logical consistency, parameter traceability, knowledge reuse rate, environmental compliance, and parameter compliance of the schemes. This method not only promoted the standardization and efficiency of construction technology disclosure activities for deep foundation pit projects but also enhanced the visualization and intelligence level of the schemes. Full article

Over recent years, driven by intertwined economic, social, environmental, and technological factors, urbanization has accelerated at an unprecedented pace, posing complex challenges to metropolitan transport systems. This has intensified the demand for innovative mobility solutions, notably Mobility as a Service (MaaS), which promotes a paradigm shift from private vehicle ownership to mobility consumed as a service. With rapid advances in digital technologies, MaaS has gained substantial momentum, attracting significant scholarly attention for its potential to enable intelligent and sustainable transportation systems. This study aims to provide a comprehensive conceptual foundation of MaaS and its components, and to systematically examine how artificial intelligence (AI), machine learning (ML), and big data techniques are applied in this domain. Following PRISMA guidelines, a bibliometric and systematic review was conducted on peer-reviewed articles published between 2020 and 2024 and indexed in the Scopus and Web of Science databases. The analysis classifies AI applications across four MaaS integration levels: basic, intermediate, advanced, and full integration. The results show that machine learning and basic optimization dominate at the basic level; blockchain and big data are most prominent at the advanced and full levels; and deep learning is applied across all levels, with a particularly strong presence at the advanced stage for real-time, personalized mobility solutions. The findings also indicate that while most implementations focus on developed countries, there is substantial potential for adaptation in emerging markets. The paper concludes by discussing key challenges in regulatory compliance, inclusivity, and the protection of sensitive user data, and outlines future research avenues for building socially equitable, intelligent, and sustainable MaaS ecosystems. Full article

Personalized learning environments increasingly rely on learner modeling techniques that integrate both explicit and implicit data sources. This study introduces a hybrid profiling methodology that combines psychometric data from an extended Felder–Silverman Learning Style Model (FSLSM) questionnaire with behavioral analytics derived from Moodle Learning Management System interaction logs. A structured mapping process was employed to associate over 200 unique log event types with FSLSM cognitive dimensions, enabling dynamic, behavior-driven learner profiles. Experiments were conducted across three datasets: a university dataset from the International Hellenic University, a public dataset from Kaggle, and a combined dataset totaling over 7 million log entries. Deep learning models including a Sequential Neural Network, BiLSTM, and a pretrained MLSTM-FCN were trained to predict student performance across regression and classification tasks. Results indicate moderate predictive validity: binary classification achieved practical, albeit imperfect accuracy, while three-class and regression tasks performed close to baseline levels. These findings highlight both the potential and the current constraints of log-based learner modeling. The contribution of this work lies in providing a reproducible integration framework and pipeline that can be applied across datasets, offering a realistic foundation for further exploration of scalable, data-driven personalization. Full article

Background: Skin cancer represents one of the most prevalent malignancies worldwide, with melanoma accounting for approximately 75% of skin cancer-related deaths despite comprising fewer than 5% of cases. Early detection dramatically improves survival rates from 14% to over 99%, highlighting the urgent need for accurate and accessible diagnostic tools. While deep learning has shown promise in dermatological diagnosis, existing approaches lack clinical explainability and deployable interfaces that bridge the gap between research innovation and practical healthcare applications. Methods: This study implemented a comprehensive multimodal deep learning framework using the HAM10000 dataset (10,015 dermatoscopic images across seven diagnostic categories). Three CNN architectures (DenseNet-121, EfficientNet-B3, ResNet-50) were systematically compared, integrating patient metadata, including age, sex, and anatomical location, with dermatoscopic image analysis. The first implementation of XRAI (eXplanation with Region-based Attribution for Images) explainability for skin lesion classification was developed, providing spatially coherent explanations aligned with clinical reasoning patterns. A deployable web-based clinical interface was created, featuring real-time inference, comprehensive safety protocols, risk stratification, and evidence-based cosmetic recommendations for benign conditions. Results: EfficientNet-B3 achieved superior performance with 89.09% test accuracy and 90.08% validation accuracy, significantly outperforming DenseNet-121 (82.83%) and ResNet-50 (78.78%). Test-time augmentation improved performance by 1.00 percentage point to 90.09%. The model demonstrated excellent performance for critical malignant conditions: melanoma (81.6% confidence), basal cell carcinoma (82.1% confidence), and actinic keratoses (88% confidence). XRAI analysis revealed clinically meaningful attention patterns focusing on irregular pigmentation for melanoma, ulcerated borders for basal cell carcinoma, and surface irregularities for precancerous lesions. Error analysis showed that misclassifications occurred primarily in visually ambiguous cases with high correlation (0.855–0.968) between model attention and ideal features. The web application successfully validated real-time diagnostic capabilities with appropriate emergency protocols for malignant conditions and comprehensive cosmetic guidance for benign lesions. Conclusions: This research successfully developed the first clinically deployable skin lesion classification system combining diagnostic accuracy with explainable AI and practical patient guidance. The integration of XRAI explainability provides essential transparency for clinical acceptance, while the web-based deployment democratizes access to advanced dermatological AI capabilities. Comprehensive validation establishes readiness for controlled clinical trials and potential integration into healthcare workflows, particularly benefiting underserved regions with limited specialist availability. This work bridges the critical gap between research-grade AI models and practical clinical utility, establishing a foundation for responsible AI integration in dermatological practice. Full article

The wide application of natural gas will help to achieve the goal of double carbon. The potential of energy saving and carbon reduction after deep condensation of a natural gas flue gas can reach as much as 10~12%. In this paper, the interplay between sensible and latent heat transfer in the process of condensation heat transfer was explored by adjusting the flue gas temperature, relative humidity, cooling water temperature, and other parameters entering the condensation heat exchanger by the condensation heat transfer experimental platform of a gas-fired boiler. The Ln number presents the proportion of water vapor condensation in flue gas, and the P number shows the total amount of water vapor condensation. The increase in Ln and p values promotes the enhancement of water vapor condensation, and the condensation heat transfer coefficient can reach about three to eight times that of a sensible heat transfer coefficient. As the Ln number increases from 0 to 0.35, the promoting effect of sensible heat is enhanced, up to a maximum of 2.5 times. However, from 0.35 to 0.75, the promotion gradually weakens. When the Ln number exceeds 0.75, sensible heat transfer begins to be suppressed, with a minimum coefficient of 0.7. The correction term of the sensible and latent heat transfer coefficient is added to the existing empirical correlation of pure convection heat transfer, which is applicable to various heat exchanger structures and verified by experiments. The micro-element superposition calculation method of the condensing heat exchanger is proposed to realize the digital accurate design of the condensing heat exchanger, which lays the foundation for the extensive promotion and application of the flue gas condensing heat exchanger. Full article

Flying-wing aircraft based on high-aspect-ratio wings are a popular configuration for many aerospace engineering applications. Cracked (or cross) control surface structures can adjust the aerodynamic characteristics of flying-wing aircraft. Deep investigations into the effects of such a control surface can provide a helpful design foundation. This paper investigates the mass distribution influences of cracked control surfaces on gust responses of high-aspect-ratio flying wings. Validated finite element modelling, revised by detailed ground vibration test (GVT) with a frequency error of less than 10%, reveals that root boundary conditions significantly affect the natural modes and frequencies of present wings with cracked control surfaces. Changes in control surface (CS) mass have a critical impact on gust response: a 150 g increase in CS mass results in a 15–22% increase in peak response acceleration and a 25–30% increase in response duration, while redistributing mass to the outboard CS reduces the peak response by 18–26% while keeping the total mass consistent. The results can provide an effective suppression strategy for the gust responses of flying-wing configurations without redesigning the main structure. Full article

In this study, the key challenges faced by global navigation satellite systems (GNSSs) in the field of security are addressed, and an eye diagram-based deep learning framework for intelligent classification of interference types is proposed. GNSS signals are first transformed into two-dimensional eye diagrams, enabling a novel visual representation wherein interference types are distinguished through entropy-centric feature analysis. Specifically, the quantification of information entropy within these diagrams serves as a theoretical foundation for extracting salient discriminative features, reflecting the structural complexity and uncertainty of the underlying signal distortions. We designed a hybrid architecture that integrates spatial feature extraction, gradient stability enhancement, and time dynamics modeling capabilities and combines the advantages of a convolutional neural network, residual network, and bidirectional long short-term memory network. To further improve model performance, we propose an improved coati optimization algorithm (ICOA), which combines chaotic mapping, an elite perturbation mechanism, and an adaptive weighting strategy for hyperparameter optimization. Compared with mainstream optimization methods, this algorithm improves the convergence accuracy by more than 30%. Experimental results on jamming datasets (continuous wave interference, chirp interference, pulse interference, frequency-modulated interference, amplitude-modulated interference, and spoofing interference) demonstrate that our method achieved performance in terms of accuracy, precision, recall, F1 score, and specificity, with values of 98.02%, 97.09%, 97.24%, 97.14%, and 99.65%, respectively, which represent improvements of 1.98%, 2.80%, 6.10%, 4.59%, and 0.33% over the next-best model. This study provides an efficient, entropy-aware, intelligent, and practically feasible solution for GNSS interference identification. Full article

The rapid evolution of unmanned aerial vehicle (UAV) technology and low-altitude economic development have propelled drone applications in critical infrastructure monitoring, particularly in intelligent transportation systems where real-time aerial image processing has emerged as a pressing requirement. We address the pivotal challenge of highway lane extraction from low-altitude UAV perspectives by applying a novel three-stage framework. This framework consists of (1) a deep learning-based semantic segmentation module that employs an enhanced STDC network with boundary-aware loss for precise detection of roads and lane markings; (2) an optimized polynomial fitting algorithm incorporating iterative classification and adaptive error thresholds to achieve robust lane marking consolidation; and (3) a global optimization module designed for context-aware lane generation. Our methodology demonstrates superior performance with 94.11% F1-score and 93.84% IoU, effectively bridging the technical gap in UAV-based lane extraction while establishing a reliable foundation for advanced traffic monitoring applications. Full article

Deep foundation pit construction faces significant safety challenges—including frequent accidents and severe disaster consequences—due to inherent complexity and uncertainty. Conventional risk assessment methods cannot adequately evaluate these complex engineering systems. This study introduces the concept of resilience to analyze safety issues during the deep foundation pits construction process and develops a safety resilience evaluation model based on the extension cloud model theory. First, based on the characteristics of the deep foundation pit construction process and the four stages of safety resilience, a safety resilience curve for deep foundation pit construction is plotted. Then, using multi-text analysis, an evaluation indicator list for deep foundation pit construction safety resilience is constructed, comprising 4 primary indicators and 24 secondary indicators. Next, based on the extension cloud model theory, the IF-AHP and entropy weight methods are combined to calculate the cloud membership degrees, systematically constructing a safety resilience evaluation model for deep foundation pit construction. Taking the Nanchang HH Center deep foundation pit project as an example, the model’s effectiveness and accuracy are validated. The results indicate that the safety resilience level of this deep foundation pit project is Grade IV, consistent with the actual engineering conditions, thereby validating the scientific validity of this method. This study innovatively applies the concepts of safety resilience and the extension cloud model to deep foundation pit construction assessment, providing a suitable method for evaluating safety risks in deep foundation pit construction projects. The model assists decision-makers in appropriate risk classification and scientific risk prevention strategies, enhances the safety management system for deep foundation pit construction, and even promotes the sustainable development of the construction industry. Full article

Cold stress severely limits legume productivity, threatening global food security, particularly in climate-vulnerable regions. This review synthesizes advances in understanding and enhancing cold tolerance in key legumes (chickpea, soybean, lentil, and cowpea), addressing three core questions: (1) molecular/physiological foundations of cold tolerance; (2) how emerging technologies accelerate stress dissection and breeding; and (3) integration strategies and deployment challenges. Legume cold tolerance involves conserved pathways (e.g., ICE-CBF-COR, Inducer of CBF Expression, C-repeat Binding Factor, Cold-Responsive genes) and species-specific mechanisms like soybean’s GmTCF1a-mediated pathway. Multi-omics have identified critical genes (e.g., CaDREB1E in chickpea, NFR5 in pea) underlying adaptive traits (membrane stabilization, osmolyte accumulation) that reduce yield losses by 30–50% in tolerant genotypes. Technologically, AI and high-throughput phenotyping achieve >95% accuracy in early cold detection (3–7 days pre-symptoms) via hyperspectral/thermal imaging; deep learning (e.g., CNN-LSTM hybrids) improves trait prediction by 23% over linear models. Genomic selection cuts breeding cycles by 30–50% (to 3–5 years) using GEBVs (Genomic estimated breeding values) from hundreds of thousands of SNPs (Single-nucleotide polymorphisms). Advanced sensors (LIG-based, LoRaWAN) enable real-time monitoring (±0.1 °C precision, <30 s response), supporting precision irrigation that saves 15–40% water while maintaining yields. Key barriers include multi-omics data standardization and cost constraints in resource-limited regions. Integrating molecular insights with AI-driven phenomics and multi-omics is revolutionizing cold-tolerance breeding, accelerating climate-resilient variety development, and offering a blueprint for sustainable agricultural adaptation. Full article

Bupleurum is a Chinese medicinal material widely used in clinical practice. Its medicinal component is the dried roots of either the Umbrella plant Bupleurum chinense DC or Bupleurum scorzonerifolium Willd. This review systematically searched major scientific databases such as Web of Science, PubMed, and ScienceDirect, and found that it contains various bioactive substances including saikosaponins, polysaccharides, flavonoids, and volatile oils. These components have demonstrated significant efficacy in anti-tumor, anti-inflammatory, and neuroprotective activities. Research has confirmed that this medicinal herb can exert its pharmacological effects by promoting tumor cell apoptosis, inhibiting cell proliferation, regulating inflammatory signaling pathways, and alleviating neuroinflammation. Additionally, its antipyretic and antiviral properties have also garnered widespread attention. However, clinical data regarding its optimal dosage, administration routes, and safety are still insufficient, necessitating further trials for validation. Investigating the synergistic effects of Bupleurum with other drugs and the safety of its use in different populations are also key directions of current research. Given the urgent need for efficient and sustainable healthcare in modern society, a deep understanding of the mechanisms and safety of Bupleurum is of significant importance for its validation as a foundation for new drug development. In summary, Bupleurum, as a multifunctional natural product, has broad application prospects and is expected to play a greater role in future medical research and clinical practice. Full article

To foster the performance of wireless communication while saving energy, the integration of Intelligent Reflecting Surfaces (IRS) into autonomous vehicle (AV) communication networks is considered a powerful technique. This paper proposes a novel IRS-assisted vehicular communication model that combines Lagrange optimization and Gradient-Based Phase Optimization to determine the optimal transmission power, optimal interference transmission power, and IRS phase shifts. Additionally, the proposed model help increase the Signal-to-Interference-plus-Noise Ratio (SINR) by utilizing IRS, which leads to maximizes energy efficiency and the achievable data rate under a variety of environmental conditions, while guaranteeing that resource limits are satisfied. In order to represent dense vehicular environments, practical constraints for the system model, such as IRS reflection efficiency and interference, have been incorporated from multiple sources, namely, Device-to-Device (D2D), Vehicle-to-Vehicle (V2V), Vehicle-to-Base Station (V2B), and Cellular User Equipment (CUE). A Lagrangian optimization approach has been implemented to determine the required transmission interference power and the best IRS phase designs in order to enhance the system performance. Consequently, a one-dimensional convolutional neural network has been implemented for the optimized data provided by this framework as training input. This deep learning algorithm learns to predict the required optimal IRS settings quickly, allowing for real-time adaptation in dynamic wireless environments. The obtained results from the simulation show that the combined optimization and prediction strategy considerably enhances the system reliability and energy efficiency over baseline techniques. This study lays a solid foundation for implementing IRS-assisted AV networks in real-world settings, hence facilitating the development of next-generation vehicular communication systems that are both performance-driven and energy-efficient. Full article

Long Short-Term Memory (LSTM) networks have revolutionized the field of deep learning, particularly in applications that require the modeling of sequential data. Originally designed to overcome the limitations of traditional recurrent neural networks (RNNs), LSTMs effectively capture long-range dependencies in sequences, making them suitable for a wide array of tasks. This survey aims to provide a comprehensive overview of LSTM architectures, detailing their unique components, such as cell states and gating mechanisms, which facilitate the retention and modulation of information over time. We delve into the various applications of LSTMs across multiple domains, including the following: natural language processing (NLP), where they are employed for language modeling, machine translation, and sentiment analysis; time series analysis, where they play a critical role in forecasting tasks; and speech recognition, significantly enhancing the accuracy of automated systems. By examining these applications, we illustrate the versatility and robustness of LSTMs in handling complex data types. Additionally, we explore several notable variants and improvements of the standard LSTM architecture, such as Bidirectional LSTMs, which enhance context understanding, and Stacked LSTMs, which increase model capacity. We also discuss the integration of Attention Mechanisms with LSTMs, which have further advanced their performance in various tasks. Despite their strengths, LSTMs face several challenges, including high Computational Complexity, extensive Data Requirements, and difficulties in training, which can hinder their practical implementation. This survey addresses these limitations and provides insights into ongoing research aimed at mitigating these issues. In conclusion, we highlight recent advances in LSTM research and propose potential future directions that could lead to enhanced performance and broader applicability of LSTM networks. This survey serves as a foundational resource for researchers and practitioners seeking to understand the current landscape of LSTM technology and its future trajectory. Full article

This paper presents an analysis of the influence of the termination geometry of an undercutting anchor drive bolt and the shape of the bottom of the anchor hole on the initiation and progression of failure processes in a rock medium. The study employed the finite element method (FEM) to model various bolt termination configurations, including cylindrical terminations with a 2 × 2 mm chamfer, a rounded termination with radius R, and a conical termination. The interaction of these bolt geometries with both cylindrical and conical hole bottoms was analyzed. The numerical simulations enabled the identification of stress concentration zones and crack propagation paths, which are critical to understanding the efficiency and mechanism of rock failure. The results indicate that the geometry of the bolt termination significantly influences stress distribution within the contact zone, as well as the extent and morphology of the resulting failure zone. Specifically, employing a cylindrical termination with a 2 × 2 mm chamfer in combination with a conical hole bottom promotes the development of deep fractures, which may lead to the detachment of larger rock fragments. This mechanism may be useful in the development of non-explosive rock fragmentation technologies. The findings provide a foundation for further optimization of anchor designs and the development of targeted excavation methods in mining and geotechnical engineering. Full article