Search Results (10,734)

This paper presents a systematic approach for rapid assessment of the performance and robustness of linear control systems through geometric analysis in the complex plane. By combining indirect performance indices within a defined zone of desired performance in the complex s-plane, a connection is established with direct performance indices, forming a foundation for the synthesis of control algorithms that ensure root placement within this zone. Analytical relationships between the complex variables s and z are derived, thereby defining an equivalent zone of desired performance for discrete-time systems in the complex z-plane. Methods for verifying digital algorithms with respect to the desired performance zone in the z-plane are presented, along with a visual assessment of robustness through radii describing robust stability and robust performance, representing performance margins under parameter variations. Through parametric modeling of controlled processes and their projections in the complex s- and z-domains, the influence of the discretization method and sampling period, as forms of a priori uncertainty, is analyzed. This paper offers original derivations for MISO systems, facilitating the analysis, explanation, and understanding of the dynamic behavior of real-world controlled processes in both the continuous and discrete-time domains, and is aimed at integration into expert systems supporting control strategy selection. The practical applicability of the proposed methodology is related to discrete control systems in energy, electric drives, and industrial automation, where parametric uncertainty and choice of method and period of discretization significantly affect both robustness and control performance. Full article

This study presents the application of Mycelial_Net, a biologically inspired deep learning architecture, to the analysis and classification of mineral images in thin section under optical microscopy. The model, inspired by the adaptive connectivity of fungal mycelium networks, was trained on a test mineral image database to extract structural features and to classify various minerals. The performance of Mycelial_Net was evaluated in terms of accuracy, robustness, and adaptability, and compared against conventional convolutional neural networks. The results demonstrate that Mycelial_Net, properly integrated with Residual Networks (ResNets), offers superior analysis capabilities, interpretability, and resilience to noise and artifacts in petrographic images. This approach holds promise for advancing automated mineral identification and geological analysis through adaptive AI systems. Full article

Labor shortages in seedling nurseries, particularly in manual inspection and replanting, hinder operational efficiency despite advancements in automation. This study aims to develop a cost-effective, GPU-free machine vision system to automate the detection of deficient seedlings in plug trays, specifically for small-scale nursery operations. The proposed Deficiency Detection and Replanting Positioning (DDRP) machine integrates low-cost components including an Intel RealSense Depth Camera D435, Raspberry Pi 4B, stepper motors, and a programmable logic controller (PLC). It utilizes OpenCV’s Haar cascade algorithm, HSV color space conversion, and Otsu thresholding to enable real-time image processing without GPU acceleration. The proposed Deficiency Detection and Replanting Positioning (DDRP) machine integrates low-cost components including an Intel RealSense Depth Camera D435, Raspberry Pi 4B, stepper motors, and a programmable logic controller (PLC). It utilizes OpenCV’s Haar cascade algorithm, HSV color space conversion, and Otsu thresholding to enable real-time image processing without GPU acceleration. Under controlled laboratory conditions, the DDRP-Machine achieved high detection accuracy (96.0–98.7%) and precision rates (82.14–83.78%). Benchmarking against deep-learning models such as YOLOv5x and Mask R-CNN showed comparable performance, while requiring only one-third to one-fifth of the cost and avoiding complex infrastructure. The Batch Detection (BD) mode significantly reduced processing time compared to Continuous Detection (CD), enhancing real-time applicability. The DDRP-Machine demonstrates strong potential to improve seedling inspection efficiency and reduce labor dependency in nursery operations. Its modular design and minimal hardware requirements make it a practical and scalable solution for resource-limited environments. This study offers a viable pathway for small-scale farms to adopt intelligent automation without the financial burden of high-end AI systems. Future enhancements, adaptive lighting and self-learning capabilities, will further improve field robustness and including broaden its applicability across diverse nursery conditions. Full article

Nematode identification and quantification are critical for understanding their impact on agricultural ecosystems. However, traditional methods rely on specialised expertise in nematology, making the process costly and time-consuming. Recent developments in technologies such as Artificial Intelligence (AI) and computer vision (CV) offer promising alternatives for automating nematode identification and counting at scale. This work reviews the current literature on nematode detection using AI techniques, focusing on their application, performance, and limitations. First, we discuss various image analysis, machine learning (ML), and deep learning (DL) methods, including You Only Look Once (YOLO) models, and evaluate their effectiveness in detecting and classifying nematodes. Second, we compare and contrast the performance of ML- and DL-based approaches on different nematode datasets. Next, we highlight how these techniques can support sustainable agricultural practices and optimise crop productivity. Finally, we conclude by outlining the key opportunities and challenges in integrating ML and DL methods for precise and efficient nematode management. Full article

The increasing complexity of web-based attacks requires the development of more effective Web Application Firewall (WAF) systems. In this study, we extend previous work by evaluating and comparing the performance of seven machine learning models for multilabel classification of web traffic, using the ECML/PKDD 2007 dataset. This dataset contains eight classes: seven representing different types of attacks and one representing normal traffic. Building on prior experiments that analyzed Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) models, we incorporate four additional models frequently cited in the related literature: Random Forest (RF), Support Vector Machine (SVM), Gradient Boosting (GB), and Feedforward Neural Networks (NN). Each model is trained and evaluated under consistent preprocessing and validation protocols. We analyze their performance using key metrics such as accuracy, precision, recall, F1-score, and training time. The results provide insights into the suitability of each method for WAF classification tasks, with implications for real-time intrusion detection systems and security automation. This study represents the first unified multilabel evaluation of classical and deep learning approaches on the ECML/PKDD 2007 dataset, offering guidance for practical WAF deployment. Full article

The automated inspection of civil infrastructure with Unmanned Aerial Vehicles (UAVs) is hampered by the challenge of accurately segmenting multi-damage in high-resolution imagery. While foundational models like the Segment Anything Model (SAM) offer data-efficient segmentation, their effectiveness is constrained by prompting strategies, especially for geometrically complex defects. This paper presents a comprehensive comparative analysis of deep learning strategies to identify an optimal deep learning pipeline for segmenting cracks, efflorescences, and exposed rebars. It systematically evaluates three distinct end-to-end segmentation frameworks: the native output of a YOLO11 model; the Segment Anything Model (SAM), prompted by bounding boxes; and SAM, guided by a point-prompting mechanism derived from the detector’s probability map. Based on these findings, a final, optimized hybrid pipeline is proposed: for linear cracks, the native segmentation output of the SAHI-trained YOLO model is used, while for efflorescence and exposed rebar, the model’s bounding boxes are used to prompt SAM for a refined segmentation. This class-specific strategy yielded a final mean Average Precision (mAP50) of 0.593, with class-specific Intersection over Union (IoU) scores of 0.495 (cracks), 0.331 (efflorescence), and 0.205 (exposed rebar). The results establish that the future of automated inspection lies in intelligent frameworks that leverage the respective strengths of specialized detectors and powerful foundation models in a context-aware manner. Full article

In regions where seismic loads pose a significant danger to the structural stability of buildings, developing sustainable solutions for increasing the ductility of structural members is of great importance. One of the contemporary, emerging approaches is to use the greater confinement of concrete using multispiral reinforcement. A numerical model of two variants of Multispiral-Reinforced Concrete Columns (MRCCs) that differ in their axial loads using FEA was developed and validated. A non-linear combined fracture-plasticity concrete model with the crack band approach and an embedded reinforced model with bond slip were used. The main finding is that higher axial loads do not significantly increase the stiffness response, but reduce ductility (achieved drift). The achieved force agreement between the simulation and the experiment is within $2\%$ at the peak and within $24\%$ at the largest column drift in the post-peak region. For the purpose of rapid prototyping, a plugin that enables the user to quickly change various properties of MRCC geometry using an automated approach instead of modeling individual variants from zero is proposed. This overall approach was developed to both save on user time spent modeling and on the great costs that involve manufacturing and testing of real-scale specimens. Full article

Digital Twins rely on accurate geometric models to ensure reliable representation, yet maintaining and updating these models remains a persistent challenge. This paper addresses one aspect of this challenge by focusing on the verification of photogrammetry-based models. It introduces a verification framework that defines measurable data quality elements and establishes conditions to assess whether model quality is maintained, improved, or degraded. Validation through a university building case study demonstrates the framework’s ability to detect quality differences between visually similar models. Meeting only one of the three verification conditions, the new model shows quality degradation, primarily due to reduced positional accuracy and resolution, making it unsuitable to replace the previous version used in the Digital Twin. Additionally, the developed web tool prototype enables the automated calculation of the framework’s verification scores. This study contributes to the growing discussion on Digital Twin maintenance by providing practical insights for improving the reliability of geometric models. Full article

This paper introduces a versatile approach to achieve specific surface functionality with-out the need for detailed knowledge of surface topography. This is accomplished for applications targeting wettability properties by integrating contact angle measurement into a micromachining setup with a nanosecond pulsed UV laser, allowing for fully automated programs to find optimal functionalization without requiring knowledge on the topography or on possible laser-induced chemical changes itself. This study investigates the impact of various processing parameters, including laser pulse energy, scanning speed, hatching distance, jump speed, and laser repetition rate, on the wetting properties of two widely used polymers: polyethylene (PE) and ethylene propylene diene monomer (EPDM). A design of experiment (DOE) approach is used for experimental design and subsequent modeling. Finally, the effectiveness of this new approach is evaluated and compared with conventional methods. Full article

Background/Objectives: The factors contributing to residual atherosclerotic cardiovascular disease (ASCVD) risk in individuals are not fully understood, but knowledge of the specific type of dyslipoproteinemia may help further refine risk assessment. We developed a novel phenotyping and risk assessment system that may be applied automatically using standard lipid panel parameters. Methods: NHANES data collected from 37,056 individuals during 1999–2018 were used to develop a three-dimensional dyslipidemia phenotype classification system. ARIC data from 14,632 individuals were used to train and validate the risk model. Three-dimensional Cartesian coordinates were converted to spherical coordinates, which were used as features in a logistic regression model that provides a probability of ASCVD. UK Biobank data from 354,344 individuals were used to further validate and test the model. Results: Nine lipidemia phenotypes were defined based on the concentrations of HDLC, non-HDLC and TG. These phenotypes were related to the prevalence of metabolic syndrome, pooled cohort equation (PCE) score and ASCVD-free survival. A logistic regression model including age, sex and the spherical coordinates of the phenotype provided a composite risk score with predictive accuracy comparable to that of the PCEs. Conclusions: We provided an example of how a multidimensional coordinate system may be used to define a novel lipoprotein phenotyping system to examine disease associations. When applied to an ASCVD risk model, the composite spherical coordinate risk marker, which can be fully automated, provided an F1 performance score almost as good as the PCEs, which requires other risk factors besides lipids. Full article

Tensegrity structures, by definition composed of compressed members suspended in a network of tensile cables, are characterised by a high strength-to-weight ratio and the ability to undergo reversible deformations. Their application as cores of sandwich panels represents an innovative approach to lightweight design, enabling the regulation of mechanical properties while reducing material consumption. This study presents a finite element modelling procedure that combines analytical determination of prestress using singular value decomposition with implementation in the ABAQUS™ 2019 software. Geometry generation and prestress definitions were automated with Python 3 scripts, while algebraic analysis of individual modules was performed in Wolfram Mathematica. Two models were investigated: M1, composed of four identical modules, and M2, composed of four modules arranged in two mirrored pairs. Model M1 exhibited a linear elastic response with a constant global stiffness of 13.9 kN/mm, stable regardless of the prestress level. Model M2 showed nonlinear hardening behaviour with variable stiffness ranging from 0.135 to 1.1 kN/mm and required prestress to ensure static stability. Eigenvalue analysis confirmed the full stability of M1 and the increase in stability of M2 upon the introduction of prestress. The proposed method enables precise control of prestress distribution, which is crucial for the stability and stiffness of tensegrity structures. The M2 configuration, due to its sensitivity to prestress and variable stiffness, is particularly promising as an adaptive sandwich panel core in morphing structures, adaptive building systems, and deployable constructions. Full article

Background/Objectives: This study presents an artificial intelligence (AI)-supported incentive spirometry system designed to explore the feasibility of automating the monitoring of respiratory exercises, a critical nursing intervention for maintaining pulmonary function and reducing postoperative complications. Methods: This system uses a tablet’s camera to track a standard spirometer’s volume indicator in real-time, reducing the manual nursing workload, unlike traditional mechanical spirometers that lack feedback capabilities. Image processing techniques analyze exercise performance, while the interface provides instant feedback, data recording, and graphical display. Machine learning models (Random Forest, XGBoost, Gradient Boosting, SVM, Logistic Regression, KNN) were trained on scripted patient data, including demographics, smoking status, and spirometry measurements, to classify respiratory performance as “poor”, “good”, or “excellent”. Results: The ensemble methods demonstrated exceptional performance, achieving 100% accuracy and R² = 1.0, with cross-validation mean accuracies exceeding 99.4%. This feasibility study demonstrates the technical viability of this AI-driven approach and lays the groundwork for future clinical validation. Conclusions: This system presents a potential cost-effective, accessible solution suitable for both clinical and home settings, potentially integrating into standard respiratory care protocols. This system not only reduces nursing workload but also has the potential to improve patient adherence. This pilot study demonstrates the technical feasibility and potential of this AI-driven approach, laying the groundwork for future clinical validation. Full article

Background: Artificial intelligence (AI), the overarching field that includes machine learning (ML) and its subfield deep learning (DL), is rapidly transforming clinical research by enabling the analysis of high-dimensional data and automating the output of diagnostic and prognostic tests. As clinical trials become increasingly complex and costly, ML-based approaches (especially DL for image and signal data) offer promising solutions, although they require new approaches in clinical education. Objective: Explore current and emerging AI applications in oncology and cardiology, highlight real-world use cases, and discuss the challenges and future directions for responsible AI adoption. Methods: This narrative review summarizes various aspects of AI technology in clinical research, exploring its promise, use cases, and its limitations. The review was based on a literature search in PubMed covering publications from 2019 to 2025. Search terms included “artificial intelligence”, “machine learning”, “deep learning”, “oncology”, “cardiology”, “digital twin”. and “AI-ECG”. Preference was given to studies presenting validated or clinically applicable AI tools, while non-English articles, conference abstracts, and gray literature were excluded. Results: AI demonstrates significant potential in improving diagnostic accuracy, facilitating biomarker discovery, and detecting disease at an early stage. In clinical trials, AI improves patient stratification, site selection, and virtual simulations via digital twins. However, there are still challenges in harmonizing data, validating models, cross-disciplinary training, ensuring fairness, explainability, as well as the robustness of gold standards to which AI models are built. Conclusions: The integration of AI in clinical research can enhance efficiency, reduce costs, and facilitate clinical research as well as lead the way towards personalized medicine. Realizing this potential requires robust validation frameworks, transparent model interpretability, and collaborative efforts among clinicians, data scientists, and regulators. Interoperable data systems and cross-disciplinary education will be critical to enabling the integration of scalable, ethical, and trustworthy AI into healthcare. Full article

This article presents the design, implementation, and comparative evaluation of two Unmanned Aerial Vehicles (UAV)-based architectures for automated road data acquisition and processing. The first system uses Arduino Portenta H7 boards to perform real-time inference at the edge, reducing connectivity dependency. The second employs ESP32-CAM modules that transmit raw data for remote server-side processing. We experimentally validated and compared both systems in terms of power consumption, latency, and detection accuracy. Results show that the Portenta-based system consumes 36.2% less energy and achieves lower end-to-end latency (10,114 ms vs. 11,032 ms), making it suitable for connectivity-constrained scenarios. Conversely, the ESP32-CAM system is simpler to deploy and allows rapid model updates at the server. These findings provide a reference for Intelligent Transportation System (ITS) deployments requiring scalable, energy-efficient, and flexible road monitoring solutions. Full article

Precision Feeding Systems (PFS) demonstrate transformative potential in advancing sustainable and efficient production within modern animal husbandry. However, existing research lacks a synthesis of PFS applications in livestock farming and offers little targeted guidance for China’s rapidly growing rabbit industry. The objective of this review is to bridge this gap by synthesizing current knowledge on PFS technologies—including sensor networks, artificial intelligence (AI), automated controls, and data analytics—and providing a structured framework for their implementation in rabbit production. This study selects and analyzes 112 core references, establishing a foundational database for comprehensive evaluation. The key contributions of this work are threefold: first, it outlines the core components and operational mechanisms of PFS; second, it identifies major challenges such as sensor reliability in dynamic environments, data security risks, limited explainability of AI models, and interoperability barriers; and third, it proposes a customized strategy for PFS adoption in rabbit farming, emphasizing phased implementation, cross-system integration, and iterative optimization. The primary outcomes and advantages of adopting such a system include significant improvements in feed efficiency, resource utilization, animal welfare, and waste reduction—critical factors given rabbits’ sensitive digestive systems and precise nutritional needs. Furthermore, this review outlines a future research agenda aimed at developing resilient sensors, explainable AI frameworks, and multi-objective optimization engines to enhance the commercial scalability and sustainability of PFS in rabbit husbandry and beyond. Full article