Search Results (330)

The increasing complexity of modern electrical substations—driven by renewable integration, advanced automation, and asset aging—necessitates a transition from reactive maintenance toward intelligent, data-driven strategies. Predictive maintenance (PdM), supported by artificial intelligence, enables early fault detection and remaining useful life (RUL) estimation, while Digital Twin (DT) technology provides synchronized cyber–physical representations for situational awareness and risk-free validation of maintenance decisions. This study proposes a five-layer DT-enabled PdM architecture integrating standards-based data acquisition, semantic interoperability (IEC 61850, CIM, and OPC UA Part 17), hybrid AI analytics, and cyber-secure decision support aligned with IEC 62443. The framework is validated using utility-grade operational data from the SS1 substation of the Badra Oil Field, comprising approximately one million multivariate time-stamped measurements and 139 confirmed fault events across transformer, feeder, and environmental monitoring systems. Fault detection is formulated as a binary classification task using event-window alignment to the 1 min SCADA timeline, preserving realistic operational class imbalance. Five supervised learning models—a Random Forest, Gradient Boosting, a Support Vector Machine, a Deep Neural Network, and a stacked ensemble—were benchmarked, with the ensemble embedded within the DT core representing the operational predictive model. Experimental results demonstrate strong performance, achieving an F1-score of 0.98 and an AUC of 0.995. The results confirm that the proposed DT–AI framework provides a scalable, interoperable, and cyber-resilient foundation for deployment-ready predictive maintenance in modern substation automation systems. Full article

Introduction: Pneumonitis represents one of the clinically significant adverse events observed in patients with non-small-cell lung cancer (NSCLC) who receive durvalumab as consolidation therapy after chemoradiotherapy (CRT). Although clinical factors such as radiation dose (e.g., V20) and interstitial lung abnormalities (ILAs) have been reported as risk predictors, accurate and objective prognostication remains difficult. This study aimed to develop a radiomics-based machine learning model to predict grade ≥ 2 pneumonitis. Methods: This retrospective study included patients with unresectable NSCLC who received CRT followed by durvalumab. Radiomic features, including first-order and texture and shape-based features with wavelet transformation were extracted from whole-lung regions on pre-durvalumab computed tomography (CT) images. Machine learning models, support vector machines, k-nearest neighbor, neural networks, and naïve Bayes classifiers were developed and evaluated using a testing cohort. Model performance was assessed using five-fold cross-validation. Conventional predictors, including V20 and ILAs, were also assessed using logistic regression and receiver operating characteristic analysis. Results: Among 123 patients, 44 (35.8%) developed grade ≥ 2 pneumonitis. The best-performing model, a support vector machine, achieved an AUC of 0.88 and accuracy of 0.81, the conventional model showed lower performance with an AUC of 0.71 and accuracy of 0.64. Conclusions: Radiomics-based machine learning demonstrated superior performance over clinical parameters in predicting pneumonitis. This approach may enable individualized risk stratification and support early intervention in patients with NSCLC. Full article

Optimizing heat treatment processes requires an understanding of the complex relationships between compositions, processing parameters, microstructures, and properties. Traditional experimental approaches are costly and time-consuming, whereas machine learning methods suffer from critical data scarcity. In this study, gradient boosting models were developed to predict microstructural phase fractions and mechanical properties using synthetic training data generated from an established metallurgical theory. A 400-sample dataset spanning eight AISI steel grades was created based on Koistinen–Marburger martensite kinetics, the Grossmann hardenability theory, and empirical property correlations from ASM handbooks. Following systematic hyperparameter optimization via 5-fold cross-validation, gradient boosting achieved R² = 0.955 for hardness (RMSE = 2.38 HRC), R² = 0.949 for tensile strength (RMSE = 87.6 MPa), and R² = 0.936 for yield strength, outperforming the Random Forest, Support Vector Regression, and Neural Networks by 7–13%. Feature importance analysis identified the tempering temperature (38.4%), carbon equivalent (15.4%), and carbon content (13.0%) as the dominant factors. Model predictions demonstrated physical consistency with the literature data (mean error of 1.8%) and satisfied the fundamental metallurgical relationships. This methodology provides a scalable and cost-effective approach for heat treatment optimization by reducing experimental requirements based on learning curve analysis while maintaining prediction accuracy within the measurement uncertainty. Full article

Large language models (LLMs) are increasingly used as rubric-guided graders for short-answer exams, but their decisions can be unstable across prompts and vulnerable to answer-side prompt injection. In this paper, we study SteadyEval, a guardrailed exam-grading pipeline in which an adversarially trained LoRA filter (SteadyEval-7B-deep) preprocesses student answers to remove answer-side prompt injection, after which the original Mistral-7B-Instruct rubric-guided grader assigns the final score. We build two exam-grading pipelines on top of Mistral-7B-Instruct: a baseline pipeline that scores student answers directly, and a guardrailed pipeline in which a LoRA-based filter (SteadyEval-7B-deep) first removes injection content from the answer and a downstream grader then assigns the final score. Using two rubric-guided short-answer datasets in machine learning and computer networking, we generate grouped families of clean answers and four classes of answer-side attacks, and we evaluate the impact of these attacks on score shifts, attack success rates, stability across prompt variants, and alignment with human graders. On the pooled dataset, answer-side attacks inflate grades in the unguarded baseline by an average of about +1.2 points on a 1–10 scale, and substantially increase score dispersion across prompt variants. The guardrailed pipeline largely removes this systematic grade inflation and reduces instability for many items, especially in the machine-learning exam, while keeping mean absolute error with respect to human reference scores in a similar range to the unguarded baseline on clean answers, with a conservative shift in networking that motivates per-course calibration. Chief-panel comparisons further show that the guardrailed pipeline tracks human grading more closely on machine-learning items, but tends to under-score networking answers. These findings are best interpreted as a proof-of-concept guardrail and require per-course validation and calibration before operational use. Full article

Background/Objectives: Accurate and reproducible grading of lumbar spinal stenosis (LSS) is clinically critical for guiding treatment decisions and patient management, yet manual assessment remains challenging due to imaging variability and inter-observer subjectivity. To address these limitations, this study aimed to evaluate the generalizability of deep learning–based feature extraction methods—VGG19, ConvNeXt-Tiny, and DINOv2—combined with classical machine learning classifiers for automated multi-grade LSS assessment. Automated grading enables objective, reproducible, and scalable assessment of lumbar spinal stenosis severity, addressing key limitations of manual interpretation. Methods: Axial MRI images were processed using pretrained VGG19, ConvNeXt-Tiny, and DINOv2 models to extract deep features. Logistic Regression, Support Vector Machine (SVM), and LightGBM were trained on internal datasets and externally validated using MRI data from the University of Phayao Hospital. Performance was assessed using accuracy, precision, recall, F1-score, confusion matrices, and multi-class ROC curves. Results: VGG19-based features yielded the strongest external performance, with Logistic Regression achieving the highest accuracy (0.9556) and F1-score (0.9558). External validation further demonstrated excellent discrimination, with AUC values ranging from 0.994 to 1.000 across all severity grades. SVM (0.9333 accuracy) and LightGBM (0.9222 accuracy) also performed well. ConvNeXt-Tiny showed stable cross-model performance, while DINOv2 features exhibited reduced generalizability, especially with LightGBM (accuracy 0.6222). Most classification errors occurred between adjacent grades. Conclusions: Deep convolutional features—particularly VGG19—combined with classical machine learning classifiers provide robust and generalizable LSS grading across external MRI data. Despite advances in modern architectures, CNN-based feature extraction remains highly effective for spinal imaging and represents a practical pathway for clinical decision support. Full article

Background: Carotid plaque composition is central to stroke risk, but some aspects of plaque characterization are derived from ex vivo imaging, while clinical decision-making relies on in vivo ultrasound (US) and computed tomography (CT). High correlation of clinical in vivo and ex vivo imaging is necessary when including ex vivo plaque features in artificial intelligence (AI) models, but the extent of this correlation between CT and US remains poorly understood. Methods: Patients undergoing carotid endarterectomy (n = 188) were enrolled. Preoperative carotid US (n = 182) and CT (n = 156) were performed. Plaque specimens from 187 patients were imaged on ex vivo CT and US. Quantitative metrics included plaque volumes, relative calcified/non-calcified volumes, HU and grayscale distributions, Agatston and calcification scores, and heterogeneity indices (coefficient of variation). Qualitative US parameters (echogenicity, juxtaluminal echolucency, discrete white areas) were visually graded. Correlation between in vivo and ex vivo imaging was assessed, and agreement was quantified for parameters with the highest correlation with Bland–Altman analysis. Results: CT of patients and ex vivo CT showed moderate to strong correlation for total, calcified, and non-calcified plaque volumes and whole-plaque mean HU (r = 0.55–0.79; CCC = 0.43–0.74). Agatston and calcification scores correlated strongly (r = 0.78–0.80; CCC = 0.63–0.76). In contrast, most non-calcified and heterogeneity metrics showed negligible-to-weak correlation. Correlations between in vivo and ex vivo US were substantially weaker (maximum correlation: 75th grayscale percentile r = 0.35). In vivo CT overestimated calcified volume (bias: 8.7%) and in vivo US underestimated the 75th grayscale quantile (bias: −25.5 grayscale). Conclusions: Quantitative CT metrics—particularly relative calcified plaque volume and calcium scores—translate reasonably well from ex vivo to in vivo imaging and represent robust candidates for radiomics and AI-based stroke risk models, even ex vivo. Ultrasound parameters show limited translational validity, underscoring the need for volumetric clinical US and discouraging the inclusion of ex vivo ultrasound features for machine learning applications. Full article

Foodborne pathogens remain a major public health concern, and the early identification of unsafe conditions is essential for preventive control. Routine inspections generate rich textual and structured data that can support real-time assessment of pathogen-related risk. The objective of this study is to develop an explainable transformer-based framework for predicting food safety inspection grades using multimodal inspection data. We combine structured metadata with unstructured deficiency narratives and evaluate classical machine learning models, deep learning architectures, and transformer models. RoBERTa achieved the highest performance (F1 = 0.96), followed by BiLSTM (F1 = 0.95) and LightGBM (F1 = 0.92). SHapley Additive exPlanations (SHAP) analysis revealed linguistically meaningful indicators of pathogen-related hazards such as temperature abuse, pests, and unsanitary practices. The findings demonstrate that transformer-based models, combined with explainable AI (XAI), can support pathogen-oriented monitoring and real-time risk assessment. This study highlights the potential of multimodal AI approaches to enhance inspection efficiency and strengthen public health surveillance. Full article

Despite the agricultural sector’s consistently high injury rates, formal reporting is often limited, leading to sparse national datasets that hinder effective safety interventions. To address this, our study introduces a comprehensive framework leveraging advanced ensemble machine learning (ML) models to predict and interpret the severity of agricultural injuries. We use a unique, manually curated dataset of over 2400 agricultural incidents from AgInjuryNews, a public repository of news reports detailing incidents across the United States. We evaluated six ensemble models, including Gradient Boosting (GB), eXtreme Grading Boosting (XGB), Light Gradient Boosting Machine (LightGBM), Adaptive Boosting (AdaBoost), Histogram-based Gradient Boosting Regression Trees (HistGBRT), and Random Forest (RF), for their accuracy in classifying injury outcomes as fatal or non-fatal. A key contribution of our work is the novel integration of explainable artificial intelligence (XAI), specifically SHapley Additive exPlanations (SHAP), to overcome the “black-box” nature of complex ensemble models. The models demonstrated strong predictive performance, with most achieving an accuracy of approximately 0.71 and an F1-score of 0.81. Through global SHAP analysis, we identified key factors influencing injury severity across the dataset, such as the presence of helmet use, victim age, and the type of injury agent. Additionally, our application of local SHAP analysis revealed how specific variables like location and the victim’s role can have varying impacts depending on the context of the incident. These findings provide actionable, context-aware insights for developing targeted policy and safety interventions for a range of stakeholders, from first responders to policymakers, offering a powerful tool for a more proactive approach to agricultural safety. Full article

This paper establishes an integrated framework combining self-induced vibration measurements with deep learning for vibration-based remaining useful life (RUL) prediction of mechanical frame structures in mobile robots. The main innovations comprise (1) a self-induced vibration excitation system that utilizes the robot’s drive wheels to generate controlled mechanical oscillations, using a five-sensor micro-electro-mechanical system (MEMS) accelerometer array to capture non-uniform vibration mode shapes across the robot’s structure, and (2) a processing pipeline for RUL prediction using accelerometer data and early feature fusion in two machine-learning models (long short-term memory (LSTM) and a convolutional neural network (CNN)). Our research methodology includes (i) modal analysis to identify the robot’s natural frequencies, (ii) verification platform evaluation, comparing low-cost MEMS accelerometers against a reference integrated electronic piezoelectric (IEPE) accelerometer, demonstrating industrial-grade measurement quality (coherence > 98%, uncertainty 4.79–7.21%), and (iii) data-driven validation using real data from the mechanical frame, showing that the LSTM model outperforms the CNN with a 2.61× root-mean-square error (RMSE) improvement (R² = 0.99). Our solution demonstrates that early feature fusion provides sufficient information to model degradation and detect faults early at a lower cost, offering a feasible alternative to classical maintenance procedures through combined hardware validation and lightweight software suitable for Industrial Internet-of-Things (IIoT) deployment. Full article

Electric vehicles rely on regenerative braking as a means of improving energy efficiency and extending driving range. However, the optimization of torque distribution between regenerative and mechanical braking remains a challenging aspect. This study investigates machine learning techniques for predicting braking torque in light EVs with a view to improving energy recovery and reducing mechanical brake usage. For this purpose, a simulation model was developed in MATLAB/Simulink to generate a data set of 113,622 points based on speed, acceleration, road grade, vehicle weight, and road condition. Four supervised ML algorithms—Linear Regression, K-Nearest Neighbors, Decision Tree, and Random Forest—were trained and evaluated using R², MSE, RMSE, and MAE metrics. To verify the results under WLTP Class 1 driving conditions, a test was conducted on a hardware test platform for the best model. The findings indicate that Random Forest achieved the highest level of accuracy with an R² value of 0.97 in the simulation and an R² value of 0.98 in the experimental validation. These findings support the hypothesis that ML-based torque prediction is a promising approach for real-time EV braking control. Also, this study supports sustainable transportation by improving energy recovery and reducing environmental impact through advanced AI-based braking strategies. Full article

Facial palsy (FP) impairs voluntary control of facial muscles, resulting in facial asymmetry and difficulties in emotional expression. Traditional assessment methods to define the severity of FP (e.g., House–Brackmann score, HB) rely on visual examinations and, therefore, are highly examiner-dependent. This study proposes an alternative approach using facial surface electromyography (EMG) for automated HB prediction. Time-domain EMG features were extracted during different facial movements (i.e., smile, close eyes, and raise forehead) and analyzed through nine different machine learning (ML) models in 58 subjects (51.98 ± 1.67 years, 20 male) with variable facial nerve function (HB 1: n = 16, HB 2–3: n = 32; HB 4–6: n = 10). Model performances were evaluated based on accuracy, precision, recall, and F1-score. Among the evaluated models, ensemble-based approaches—particularly a random forest model with 100 trees and a decision tree ensemble—proved to be the most effective with classification accuracies ranging from 81.7 to 84.8% and from 81.7 to 84.7%, depending on the evaluated facial movement. The results indicate that ensemble-based ML models can reliably distinguish between different FP grades using non-invasive EMG data. The approach offers a robust alternative to subjective clinical scoring, potentially improving diagnostic consistency and supporting longitudinal monitoring in clinical and research applications. Full article

The increasing demand for sustainable construction materials has motivated the optimization of mortar mix designs to reduce cement consumption and its environmental impact while maintaining adequate mechanical performance. This study develops a machine learning (ML) model for optimizing mortar mixtures used in concrete roofing tiles by integrating aggregate particle packing techniques with non-linear regression algorithms, using an industry-grade dataset generated in the Central Laboratory of Wienerberger Ltd. Unlike most previous studies, which mainly focus on compressive strength, this research targets the transverse strength of industrial roof tile mortar. The proposed approach combines Tarantula Curve gradation limits, experimentally derived packing density ($\eta$), and ML regression within a unified and application-oriented workflow, representing a research direction rarely explored in the literature for optimizing concrete mix transverse strength. Fine concrete aggregates were characterized through a sand sieve analysis and subsequently adjusted according to the Tarantula Curve method to optimize packing density and minimize void content. Physical properties of cements and fine aggregates were assessed, and granulometric mixtures were evaluated using computational methods to calculate fineness modulus summation (FMS) and packing density. Mortar samples were tested for transverse strength at 1, 7, and 28 days using a three-point bending test, generating a robust dataset for modeling training. Three ML models—Random Forest Regressor (RFR), XG-Boost Regressor (XGBR), and Support Vector Regressor (SVR)—were evaluated, confirming their ability to capture nonlinear relationships between mix parameters and transverse strength. The analysis of input variables, which consistently ranked as the highest contributors according to impurity-based and permutation-based importance metrics, revealed that the duration of curing, density, and the summation of the fineness modulus significantly influenced the estimated transverse strength derived from the models. The integration of particle size distribution optimization and ML demonstrates a viable pathway for reducing cement content, lowering costs, and achieving sustainable mortar mix designs in the tile manufacturing industry. Full article

Linear cultural heritage is characterized by complex cross-regional and multi-level features, facing severe challenges of spatial resource fragmentation and an imbalance in cultural and tourism functions. However, existing research lacks quantitative analysis regarding the non-linear driving mechanisms of spatial distribution and the misalignment of culture–tourism coupling. In this study, we construct an integrated identification–explanation–coupling–governance (IECG) theoretical framework. Taking The Great Tea Road (Jinzhong Section) as a case study, our framework integrates the CCSPM, XGBoost-SHAP machine learning interpreter, and Geodetector to systematically quantify the spatial structure of heritage and the level of culture–tourism integration. The results indicate that, (1) in terms of spatial patterns, the study area exhibits an unbalanced agglomeration characteristic of “dual-primary and dual-secondary cores,” with high-density areas showing significant orientation along rivers and roads; (2) regarding driving mechanisms, the machine learning model reveals a significant “non-linear threshold effect,” with 83% of driving factors (e.g., elevation and distance to transportation) exhibiting non-linear fluctuations in their influence on heritage distribution; and, (3) in terms of culture–tourism coupling, the overall coupling coordination degree (CCD) is low (mean 0.38), indicating significant “resource–facility” spatial misalignment. The modern number of public cultural facilities (NCF) is identified as the primary obstacle restricting the transformation of high-grade heritage into tourism products. Based on these findings, we propose adaptive zoning governance strategies. This research not only theoretically clarifies the complexity of the social–ecological system of linear heritage but also provides a generalizable quantitative method for the digital protection and sustainable tourism planning of cross-regional cultural heritage. Full article

RNA In Situ Hybridization (RISH) is a powerful tool for spatial gene expression analysis, yet its quantitative use remains limited by the high cost and inaccessibility of commercial software, particularly in under-resourced settings. This study developed an Artificial Intelligence/Machine Learning (AI/ML)-assisted RISH quantification pipeline to evaluate expression patterns of High Mobility Group AT Hook-2 (HMGA2) in prostate cancer (PCa), focusing on racial disparities. We created a machine learning model capable of analyzing RISH images. Expressions of full-length (wild-type) and truncated HMGA2 isoforms were assessed in tissues from 85 men of African descent, European American, and Asian descent. A training dataset was generated for supervised learning analysis of the full cohort. RISH findings revealed that the wild-type HMGA2 isoform was significantly more abundant in tumors from men of African descent and positively correlated with increasing Gleason grade. The truncated isoform was less abundant and did not display a consistent expression pattern across racial groups. These results demonstrate the feasibility of AI/ML-based RISH quantification and suggest that elevated wild-type HMGA2 expression may represent a biomarker linked to prostate cancer aggressiveness and racial disparities. These findings highlight the importance of interdisciplinary collaboration and equitable computational tools in advancing biomarker discovery and addressing cancer health inequities. Full article

Phalaenopsis orchids are one of Taiwan’s key floral export products, and maintaining consistent quality is crucial for international competitiveness. To improve production efficiency, many orchid farms outsource the early flask seedling stage to contract growers, who raise the plants to the 2.5-inch potted seedling stage before returning them for further greenhouse cultivation. Traditionally, the quality of these outsourced seedlings is evaluated manually by inspectors who visually detect defects and assign quality grades based on experience, a process that is time-consuming and subjective. This study introduces a smart image-based deep learning system for automatic quality grading of Phalaenopsis potted seedlings, combining computer vision, deep learning, and machine learning techniques to replace manual inspection. The system uses YOLOv8 and YOLOv10 models for defect and root detection, along with SVM and Random Forest classifiers for defect counting and grading. It employs a dual-view imaging approach, utilizing top-view RGB-D images to capture spatial leaf structures and multi-angle side-view RGB images to assess leaf and root conditions. Two grading strategies are developed: a three-stage hierarchical method that offers interpretable diagnostic results and a direct grading method for fast, end-to-end quality prediction. Performance comparisons and ablation studies show that using RGB-D top-view images and optimal viewing-angle combinations significantly improve grading accuracy. The system achieves F1-scores of 84.44% (three-stage) and 90.44% (direct), demonstrating high reliability and strong potential for automated quality assessment and export inspection in the orchid industry. Full article