Search Results (2,483)

Detecting underwater debris is important for monitoring the marine environment but remains challenging due to poor image quality, visual noise, object occlusions, and diverse debris appearances in underwater scenes. This study proposes UDD-YOLO, a novel detection framework that, for the first time, applies a diffusion-based model to underwater image enhancement, introducing a new paradigm for improving perceptual quality in marine vision tasks. Specifically, the proposed framework integrates three key components: (1) a Cold Diffusion module that acts as a pre-processing stage to restore image clarity and contrast by reversing deterministic degradation such as blur and occlusion—without injecting stochastic noise—making it the first diffusion-based enhancement applied to underwater object detection; (2) an AMC2f feature extraction module that combines multi-scale separable convolutions and learnable normalization to improve representation for targets with complex morphology and scale variation; and (3) a Unified-IoU (UIoU) loss function designed to dynamically balance localization learning between high- and low-quality predictions, thereby reducing errors caused by occlusion or boundary ambiguity. Extensive experiments are conducted on the public underwater plastic pollution detection dataset, which includes 15 categories of underwater debris. The proposed method achieves a mAP50 of 81.8%, with 87.3% precision and 75.1% recall, surpassing eleven advanced detection models such as Faster R-CNN, RT-DETR-L, YOLOv8n, and YOLOv12n. Ablation studies verify the function of every module. These findings show that diffusion-driven enhancement, when coupled with feature extraction and localization optimization, offers a promising direction for accurate, robust underwater perception, opening new opportunities for environmental monitoring and autonomous marine systems. Full article

Current functional safety mechanisms mainly control the access points and perimeters of manufacturing cells without guaranteeing the integrity of their internal components or the absence of unauthorized humans or objects. In this work, we present a novel deep learning (DL)-based safety system that enhances the safety circuit designed according to functional safety principles, detecting, with great reliability, the presence of persons within the cell and, with high precision, anomalous elements of any kind. Our approach follows a two-stage DL methodology that combines contrastive learning with Bayesian clustering. First, a supervised contrastive scheme learns the characteristics of safe scenarios and distinguishes them from unsafe ones caused by workers remaining inside the cell. Next, a Bayesian mixture models the latent space of safe scenarios, quantifying deviations and enabling the detection of previously unseen anomalous objects without any specific fine-tuning. To further improve robustness, we introduce an ensemble-based hybrid latent-space methodology that maximizes performance regardless of the underlying encoders’ characteristics. The experiments are conducted on a real dataset captured in a belt-picking cell in production. The proposed system achieves 100% accuracy in distinguishing safe scenarios from those with the presence of workers, even in partially occluded cases, and an average area-under-the-curve of 0.9984 across seven types of anomalous objects commonly found in manufacturing environments. Finally, for interpretability analysis, we design a patch-based feature-ablation framework that demonstrates the model’s reliability under uncertainty and the absence of learning biases. The proposed technique enables the deployment of an innovative high-performance safety system that, to our knowledge, does not exist in the industry. Full article

Permanent Magnet Synchronous Motors (PMSMs) exhibit inherent symmetry in their electromagnetic structure yet behave as nonlinear and strongly coupled systems that are susceptible to internal parameter perturbations and external disturbances, posing challenges to effective control under dynamic operating conditions. To address these issues, this paper proposes a sliding mode control strategy for PMSMs that integrates a Novel Adaptive Reaching Law (NARL) and an Improved Terminal Fuzzy Sliding Mode Disturbance Observer (IFTSMDO), denoted as SMC-NARL-IFTSMDO. The NARL is designed with a state-dependent dynamic gain adjustment mechanism and terminal attractive factor characteristics: it increases the gain to ensure fast convergence when the system state is far from the sliding mode surface, and adaptively attenuates the gain to suppress chattering when approaching the sliding mode surface, thereby balancing the contradiction between convergence speed and chattering in traditional sliding mode control. The IFTSMDO constructs a composite sliding mode surface incorporating error derivatives, terminal power terms, and saturation functions, which enhances the sensitivity of disturbance estimation in the small-error stage, avoids high-frequency chattering caused by sign functions, and provides accurate feedforward compensation for the speed loop controller to improve the system’s anti-disturbance capability. Additionally, the asymptotic stability of the proposed control strategy is strictly proven using the Lyapunov stability theory, laying a solid theoretical foundation for its application. Experiments are conducted on a TMS320F28379D DSP-based platform, and quantitative results show that compared with the traditional sliding mode control (SMC-TRL), the proposed strategy reduces the no-load startup response time by 60%, the steady-state speed fluctuation by 60%, and the speed fluctuation under load disturbance by 81.5%, fully demonstrating its superiority in dynamic response and anti-disturbance performance. Full article

This paper addresses the challenge of ensuring quality control for electronic control units (ECUs) in the automotive industry. Traditional hardware-in-the-loop (HIL) systems, used for software quality assurance, are costly and not always readily available, leading to potential delays in software verification and release. This paper investigates the factors causing these delays and proposes process task workflow optimization using a hybrid testing approach. This approach allows software developers to conduct early integration and system testing for embedded software, complementing the formal validation stage. To support this, the paper presents a novel piece of test equipment called TestBench, which utilizes inexpensive embedded hardware like the Arduino Mega2560, which can carry out the fundamental tasks needed to test the ECU. TestBench enables local testing, remote accessibility, and the monitoring of key parameters, including voltage current consumption, and the communication bus. It also facilitates fault injection for the evaluation of communication protocol robustness. By enabling earlier and more frequent testing, TestBench aims to enhance the quality of software developer outputs and the overall quality of ECU-embedded software. The system has the potential to significantly improve the testing process, making advanced testing capabilities more accessible and cost-effective for engineering teams and educational institutions. Full article

Background/Objectives: Malignant pleural mesothelioma (MPM) is a rare, aggressive tumor with a poor prognosis and complex diagnostic pathways. Pulmonologists often play a central role in its initial recognition and investigation. This narrative review synthesizes the current evidence on the diagnostic approach to MPM, with emphasis on imaging, tissue sampling, histopathology, and emerging diagnostic innovations relevant to clinical pulmonology. Methods: English-language studies published between January 2005 and June 2025 were identified from PubMed and Scopus. International guidelines and consensus documents were also reviewed to provide an updated overview of diagnostic strategies. Results: Diagnosis of MPM relies on a stepwise integration of clinical, radiological, and pathological information. Thoracic ultrasound, computed tomography, positron emission computed tomography and magnetic resonance imaging complement each other across different stages of the diagnostic pathway. Image-guided pleural biopsy and medical thoracoscopy remain the gold standard for tissue confirmation, supported by immunohistochemistry and molecular testing. The 2021 World Health Organization classification of pleural tumors and the International Association Study of Lung Cancer 9th Edition Tumour-Node-Mestastatis system have refined histologic and staging criteria, thereby improving reproducibility and prognostic accuracy. Emerging tools, including liquid biopsy, novel serum and molecular biomarkers, artificial-intelligence-based radiomics, and breathomics, offer promise for earlier and less invasive diagnosis but require prospective validation. Conclusions: Current advances are redefining MPM diagnosis toward integrated, multidisciplinary, and precision-based models. Future priorities include standardizing diagnostic algorithms, validating minimally invasive biomarkers, and integrating AI and molecular profiling into clinical workflows to enhance patient stratification. Full article

Dental caries is one of the most widespread chronic infectious diseases for humans. It results in localized destruction of dental hard tissues and has negative impacts on systemic health. Aims: This study aims to design, model, and control a novel 4-DOF dental robotic manipulator, Dentatron, specifically tailored for dental applications. The objectives were to (1) develop a compact robotic arm optimized for dental workspace constraints, (2) implement and compare three controllers—Computed Torque Control (CTC), Fuzzy Logic Control (FLC), and Neural Network Adaptive Control (NNAC), (3) evaluate tracking accuracy, transient response, and robustness in step and trajectory tasks, and (4) assess the potential of adaptive neural controllers for future clinical integration. Materials and Methods: The Dentatron system integrates a custom-designed robotic manipulator with adaptive controllers. The methodology consists of five main stages: robot modeling, control design, neural network adaptation, training, and evaluation. Simulations were performed to evaluate performance across joint tracking and Cartesian trajectory tasks using MATLAB 2022. Human-inspired trajectory design is fundamental to the Dentatron control and simulation framework to emulate the continuous curvature and minimum jerk characteristics of human upper-limb motion. The desired end-effector paths were formulated using fifth-degree polynomial trajectories that produce bell-shaped velocity profiles with gradual acceleration changes. Results: The study revealed that the Neural Network Adaptive Controller (NNAC) achieved the fastest convergence and lowest tracking error (<3 mm RMSE), consistently outperforming Fuzzy Logic Control (FLC) and Computed Torque Control (CTC). NNAC consistently provided precise joint tracking with minimal overshoot, while FLC ensured smoother but slower responses, and CTC exhibited large overshoot and persistent oscillations, requiring precise modeling to remain competitive. Conclusion: NNAC demonstrated the most robust and accurate control performance, highlighting its promise for safe, precise, and clinically adaptable robotic assistance in dentistry. Dentatron represents a step toward the development of compact dental robots capable of enhancing the precision and efficiency of future dental procedures. Full article

AI-generated phishing emails present a growing cybersecurity threat, exploiting human psychology with high-quality, context-aware language. This paper introduces a novel two-stage detection framework that combines deep learning with psychological analysis to address this challenge. A new dataset containing 2995 GPT-o1-generated phishing emails, each labelled with Cialdini’s six persuasion principles, is created across five organisational sectors—forming one of the largest and most behaviourally annotated corpora in the field. The first stage employs a fine-tuned DistilBERT model to predict the presence of persuasion principles in each email. These confidence scores then feed into a lightweight dense neural network at the second stage for final binary classification. This interpretable design balances performance with insight into attacker strategies. The full system achieves 94% accuracy and 98% AUC, outperforming comparable methods while offering a clearer explanation of model decisions. Analysis shows that principles like authority, scarcity, and social proof are highly indicative of phishing, while reciprocation and likeability occur more often in legitimate emails. This research contributes an interpretable, psychology-informed framework for phishing detection, alongside a unique dataset for future study. Results demonstrate the value of behavioural cues in identifying sophisticated phishing attacks and suggest broader applications in detecting malicious AI-generated content. Full article

The design of multi-link vehicle suspension systems, such as the 3D double-wishbone, presents a critical challenge in automotive engineering. The process constitutes a high-dimensional, nonlinearly constrained optimization problem where traditional gradient-based methods often fail by converging to suboptimal local minima. This paper introduces a novel two-stage hybrid optimization framework designed to overcome this limitation by intelligently integrating quantum-inspired and classical techniques. The methodology explicitly defines a QUBO (Quadratic Unconstrained Binary Optimization) stage and an SQP (Sequential Quadratic Programming) stage. Stage 1 addresses the complex kinematic constraint problem by formulating it as a QUBO, which is then solved using Simulated Annealing to perform a global search, guaranteeing a physically feasible starting point. Subsequently, Stage 2 takes this feasible solution and employs an SQP algorithm to perform a high-precision local refinement, tuning the geometry to meet specific performance targets for camber and caster curves. The framework successfully converged to a design with a near-zero performance objective of 7.08 × 10⁻¹⁴. The efficacy of this hybrid approach is highlighted by the dramatic improvement from the high-error initial solution found by Simulated Annealing to the final, high-precision result from the SQP refinement. We conclude that this QUBO-SQP framework is a powerful and validated methodology for solving complex, real-world engineering design problems, effectively bridging the gap between global exploration and local precision. Full article

In the rapidly evolving landscape of logistics and last-mile delivery, optimizing efficiency and minimizing costs are paramount. This paper introduces a novel heuristic approach designed to enhance the efficiency of a truck-and-drone delivery system. Our method addresses the complex challenge of coordinating the movements of a truck, which serves as a mobile depot, and an unmanned aerial vehicle (UAV or drone), which performs rapid, short-distance deliveries. Our system proposes a two-step heuristic. For truck routes, we utilized the Concorde Solver to determine the optimal path, based on real-world road distances between locations in Bacău County, Romania. This data was meticulously collected and processed as a Traveling Salesman Problem (TSP) instance with precise geographical information. Concurrently, a drone is deployed for specific deliveries, with routes calculated using the Haversine formula to determine accurate distances based on geographical coordinates. A crucial aspect of our model is the integration of the drone’s limited autonomy, ensuring that each mission adheres to its operational capacity. Computational experiments conducted on a real-world dataset including 93 localities from Bacău County, Romania, demonstrate the effectiveness of the proposed two-stage heuristic. Compared to the optimal truck-only route, the hybrid truck-and-drone system achieved up to 15.59% cost reduction and 38.69% delivery time savings, depending on the drone’s speed and autonomy parameters. These results confirm that the proposed approach can substantially enhance delivery efficiency in realistic distribution scenarios. Full article

Sensor-less adaptive optics (SLAO) using stochastic parallel gradient descent (SPGD) offers a promising solution for wavefront correction in free-space optical communication (FSOC) systems, as it eliminates the need for conventional wavefront sensors. However, the standard SPGD algorithm’s convergence speed is limited, and it is prone to becoming trapped in local extrema, especially under complex, high-dimensional wavefront distortions in large-scale and dynamic FSOC systems, hindering its use in time-sensitive, high-precision scenarios. To address these limitations, we propose a novel Modal Stage SPGD (MSSPGD) algorithm which integrates subspace optimization techniques with the traditional SPGD algorithm. By projecting the control problem onto a reduced-dimensional Zernike modal subspace and adaptively expanding controlled modes number based on performance metric, our approach decomposes the high-dimensional optimization task into a coarse to fine search optimization problem, thereby accelerating convergence speed, reducing computational complexity, and enhancing robustness against local optima. Theoretical analysis and numerical simulations demonstrate that the proposed algorithm improves convergence speed, stability, and adaptability leading to more effective mitigation of turbulence-induced degradation in critical FSOC metrics. Experimental results further show that the MSSPGD algorithm achieves an approximately 25% reduction in iteration count compared to conventional SPGD. These enhancements prove that the algorithm highly suitable for real-time SLAO in demanding high-speed FSOC systems. Full article

Efficient delivery of exogenous genetic material remains a core challenge in plant biotechnology, holding profound implications for sustainable agricultural and forestry development. Although traditional delivery methods such as Agrobacterium-mediated transformation, gene gun bombardment, and electroporation have been widely applied in plant genetic engineering, these systems exhibit limitations including species-dependent efficacy, propensity to cause plant tissue damage, low transformation efficiency, susceptibility to environmental factors. In recent years, with the advancement of nanotechnology, nanoparticle-based nucleic acid delivery systems are emerging as novel tools for applications such as novel tools for dsRNA or transgene delivery. These systems leverage the unique physicochemical properties of nanomaterials, including size-dependent phenomena, tunable surface charge, and enhanced membrane penetration capabilities, to achieve targeted delivery and stable expression of genetic payloads. Nevertheless, nanomaterial-mediated gene delivery systems for plants are still in their nascent stages, and their widespread application faces numerous challenges. This article briefly introduces traditional delivery methods, systematically reviews the applications and progress of nanoparticle-based nucleic acid delivery systems, and discusses the cross-species applicability of nanoparticles, as well as the associated biosafety concerns. We aim to offer insights for tackling the prevailing technical bottlenecks and to provide guidance for the rational design of nanomaterials that efficiently traverse the plant cell wall–plasma membrane barrier and stably deliver nucleic acids without eliciting phytotoxicity. Full article

Background/Objectives: Alopecia areata (AA) is a common, noncicatricial autoimmune hair loss disorder characterized by relapsing inflammation and breakdown of hair follicle immune privilege. Increasing amounts of evidence suggest that pyroptosis, a lytic and inflammatory form of programmed cell death mediated by gasdermins and inflammasome activation, may play a role in AA pathogenesis. This review aims to synthesize current data on the molecular mechanisms linking inflammasome-driven pyroptosis with AA and to highlight emerging therapeutic opportunities. Methods: A comprehensive literature review was conducted focusing on mechanistic studies, ex vivo human scalp models, murine AA models, and interventional clinical data. A structured system of Levels of Evidence (LoE) and standardized nomenclature for experimental models was applied to ensure transparency in evaluating the role of pyroptosis and treatment strategies in AA. Results: Available evidence indicates that outer root sheath keratinocytes express functional inflammasome components, including NOD-like receptor family, pyrin domain containing 3 (NLRP3), adaptor-apoptosis-associated-speck-like protein (ASC), and caspase-1, and contribute to interleukin (IL)-1β release and pyroptotic cell death. Mitochondrial dysfunction, mediated by regulators such as PTEN and PINK1, amplifies NLRP3 activation and cytokine secretion, linking mitophagy impairment with follicular damage. Animal and human biopsy studies confirm increased inflammasome activity in AA lesions. Therapeutic approaches targeting pyroptosis include Janus kinase (JAK) inhibitors, biologics, Phosphodiesterase 4 (PDE4) inhibitors, mesenchymal stem cell therapy, natural compounds, and inflammasome inhibitors such as MCC950. While some agents demonstrated efficacy in clinical trials, most strategies remain at preclinical or early clinical stages. Conclusions: Pyroptosis represents a critical mechanism driving hair follicle structural and functional disruption and immune dysregulation in AA. By integrating evidence from molecular studies, disease models, and early clinical data, this review underscores the potential of targeting inflammasome-driven pyroptosis as a novel therapeutic strategy. Full article

Background/Objectives: Large language models (LLMs) have been proposed as a means of converting unstructured electronic medical records (EMRs) into structured datasets. However, concerns regarding the reliability of these models in non-English clinical text and their capacity to generate novel insights remain unresolved. We aimed to utilize an LLM to identify a hypothetical “Luminal B poor-prognosis” breast cancer subgroup (LPP) based on progesterone receptor (PR), the Ki-67 proliferation index, and grade characteristics, while concurrently validating the LLM’s accuracy. Methods: We retrospectively compiled the EMRs on 7756 female breast cancer patients from five Moscow oncology centers. An LLM with a domain-engineered prompt extracted eight clinicopathological variables (Ki-67, estrogen receptor (ER)/PR Allred status, HER2 status, grade, relapse dates, and multiple primaries). The accuracy of the model was validated in 366 randomly sampled cases against oncologist annotations using Intraclass Correlation Coefficient (ICC) and weighted κ. Following data post-processing, the complete-case cohort (n = 2347) and the HR+/HER2− stage I–III sub-cohort (n = 1419) were analyzed. Survival was estimated with Kaplan–Meier/log-rank and modeled with Cox regression (adjusted for age, stage, and treatment). Ki-67 was modeled continuously; prespecified LPP definitions were compared. Results: LLM–human agreement was high (Ki-67 ICC = 0.882; grade κ = 0.887; ER κ = 0.997; PR κ = 0.975; HER2 κ = 0.935). Date extraction was characterized by a high degree of missing data. In HR+/HER2− stage I–III disease, ER < 5 was non-prognostic; however, PR < 4 and Ki-67 ≥ 40% were indicative of inferior survival (HR 2.25 and 1.85). The most effective LPP definition (PR < 4 and Ki-67 ≥ 40%) identified a subgroup (~5.3%) of patients with markedly poorer outcomes (age, stage, and treatment adjusted HR 2.60, 95% CI 1.53–4.43) compared to the Luminal B (HER2−) subgroup. Conclusions: The developed LLM has demonstrated the ability to reliably structure non-English EMRs and enable discovery of clinically meaningful subgroups. The discovered LPP phenotype defines a small, high-risk subset warranting external validation. Given the retrospective, single-system design of the study, it is imperative to interpret the discovered phenotype features as hypothesis-generating, rather than as definitive evidence. Full article

This study presents a novel methodology for the detection and monitoring of changes in surface water bodies, with a particular emphasis on the near-eastern region of Antioquia, Colombia. The proposed approach integrates remote sensing and artificial intelligence techniques through the fusion of multi-source imagery, specifically Synthetic Aperture Radar (SAR) and optical data. The framework is structured in several stages. First, radar imagery is pre-processed using an autoencoder-based despeckling model, which leverages deep learning to reduce noise while preserving structural information critical for environmental monitoring. Concurrently, optical imagery is processed through the computation of normalized spectral indices, including NDVI, NDWI, and NDBI, capturing essential characteristics related to vegetation, water presence, and surrounding built-up areas. These complementary sources are subsequently fused into synthetic RGB composite representations, ensuring spatial and spectral consistency between radar and optical domains. To operationalize this methodology, a standardized and reproducible workflow was implemented for automated image acquisition, preprocessing, fusion, and segmentation. The Segment Anything Model (SAM) was integrated into the process to generate semantically interpretable classes, enabling more precise delineation of hydrological features, flood-prone areas, and urban expansion near waterways. This automated system was embedded in a software prototype, allowing local users to manage large volumes of satellite data efficiently and consistently. The results demonstrate that the combination of SAR and optical datasets provides a robust solution for monitoring dynamic hydrological environments, particularly in tropical mountainous regions with persistent cloud cover. The fused products enhanced the detection of small streams and complex hydrological patterns that are typically challenging to monitor using optical imagery alone. By integrating these technical advancements, the methodology supports improved environmental monitoring and provides actionable insights for decision-makers. At the local scale, municipal governments can use these outputs for urban planning and flood risk mitigation; at the regional level, environmental and territorial authorities can strengthen water resource management and conservation strategies; and at the national level, risk management institutions can incorporate this information into early warning systems and disaster preparedness programs. Overall, this research delivers a scalable and automated tool for surface water monitoring, bridging the gap between scientific innovation and operational decision-making to support sustainable watershed management under increasing pressures from climate change and urbanization. Full article

Background/Objectives: Proinsulin, the precursor to insulin, has limited activity on the insulin receptor. Proinsulin levels increase with increasing insulin resistance in type 2 diabetes due to incomplete processing by the β-cell. To assess whether the development of peptides that could convert circulating proinsulin to insulin in the blood would provide therapeutic value, we used a quantitative systems pharmacology (QSP) model of glucose homeostasis. In silico hypothesis testing such as this is an example of how modeling can inform decisions in drug discovery. Methods: In silico hypothesis testing involved (1) the addition and qualification of proinsulin biology into a preexisting QSP model, (2) the creation and validation of virtual patients (VPs) for subpopulations of type 2 diabetics based on phenotypic traits, and (3) the simulation of clinical trials evaluating the therapeutic value of the conversion of circulating proinsulin to insulin in the VPs created. Results: Proinsulin conversion led to a ~0.2% reduction in HbA1c in VPs at varying stages of diabetes, a decrease that does not hold meaningful therapeutic value. The lack of significant impact on HbA1c was likely a result of the surprisingly small effect on plasma insulin levels from proinsulin, which has a significantly slower secretion and clearance rate. Although patients with higher proinsulin/insulin ratios showed the largest reductions, clinically significant ≥ 0.5% reduction in HbA1c required ratios of proinsulin/insulin above the reported physiological range. Conclusions: This effort demonstrates how in silico hypothesis testing using QSP modeling can provide insights on the probability of success of novel interventions with minimal time and resources. These efficiencies are a means of overcoming the pressures on the pharmaceutical industry to do more with less in providing therapies that improve the lives of patients. Full article