Search Results (265)

Early identification of individuals with abnormal glucose metabolism is essential for timely intervention and prevention of diabetes-related complications. Routine laboratory testing generates large amounts of clinical data that may support automated glycemic classification through machine learning approaches. This study aimed to develop and evaluate a machine learning framework for the classification of HbA1c-defined glycemic status using routinely available clinical laboratory features. A retrospective dataset of 1434 individuals with available glycemic measurements was analyzed. Participants were categorized into HbA1c-defined normoglycemic, prediabetic-range, or diabetic-range groups. Three concurrent classification tasks were examined: HbA1c-defined dysglycemia classification, diabetic-range HbA1c classification, and multiclass HbA1c-defined glycemic-status classification. Demographic, biochemical, and hematological variables were used as predictors. Data preprocessing included missing-value handling, feature filtering, and outlier treatment. Several supervised learning algorithms were evaluated, including Logistic Regression, Random Forest, Gradient Boosting, Support Vector Machine, and Multinomial Logistic Regression. Model performance was assessed using train–test validation and cross-validation with accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve. For dysglycemia, Gradient Boosting achieved the highest AUC (0.848), while Random Forest achieved the highest accuracy (0.801) and sensitivity (0.908). For diabetic-range HbA1c, Random Forest achieved the highest AUC (0.864), whereas SVM achieved the highest accuracy (0.794). In multiclass classification, Random Forest achieved the highest accuracy (0.610), while Gradient Boosting achieved the highest macro-AUC (0.796) and macro-F1 score (0.603). Pairwise comparisons showed no statistically significant superiority of any classifier after Holm correction. Clinical-baseline and ablation analyses demonstrated that fasting glucose accounted for a substantial proportion of discrimination, with only modest incremental value from additional laboratory variables. These findings support cautious interpretation of routine laboratory-based classification models pending further validation and clinical-utility assessment. Full article

Optical coherence tomography (OCT) is a high-resolution and non-contact optical imaging and sensing modality that provides depth-resolved cross-sectional visualization of retinal microstructures. It plays an important role in the assessment of retinal diseases, including age-related macular degeneration (AMD) and diabetic macular edema (DME). However, automated OCT image classification commonly relies on fully supervised models that require large-scale expert annotations, which are costly and time-consuming because of the complex layered anatomy and subtle pathological patterns present in retinal OCT images. To reduce annotation dependence, this study proposes a self-supervised representation learning framework with local dimensionality regularization for retinal OCT image classification. The proposed method estimates the local intrinsic dimensionality of learned representations and incorporates it into an asymptotic Fisher-Rao regularization objective to mitigate local dimensional degeneration and preserve fine-grained structural information. Logarithmic scaling and geometric averaging are further introduced to reduce sensitivity to outliers and improve optimization stability. Experiments on three independent OCT datasets achieved classification accuracies of 94.35%, 92.48%, and 92.56%, respectively, demonstrating competitive performance compared with mainstream self-supervised methods. These results demonstrate that explicitly modeling local feature geometry can improve the discrimination of sensor-acquired OCT images while reducing reliance on manual annotations, providing an effective approach for intelligent analysis of biomedical optical imaging data. Full article

The rapid deployment of Internet of Things (IoT) devices across smart cities, healthcare systems, industrial automation, transportation networks, smart grids, and cyber-physical infrastructures has expanded the modern cyberattack surface. IoT devices are often constrained by limited processing capacity, memory, battery power, and communication bandwidth, making conventional security mechanisms difficult to deploy consistently at scale. Intrusion detection systems (IDSs) provide an important defensive layer; however, many machine-learning-based IDSs are developed under static assumptions and may experience performance degradation as traffic distributions evolve due to firmware changes, device onboarding, protocol updates, user behavior variation, or adaptive attacks. This paper presents a hybrid IDS framework that integrates supervised Random Forest classification, unsupervised Isolation Forest anomaly monitoring, and Kolmogorov–Smirnov (KS)-based concept drift monitoring. In the experimental pipeline, Isolation Forest is trained exclusively on benign traffic to ensure that the anomaly detector models normal behavior rather than an attack-dominated training distribution. The evaluation uses a large-scale chronologically sampled subset of the CICIoT2023 dataset containing 3,890,621 records while preserving the natural class distribution of 2.35% benign traffic and 97.65% attack traffic. The chronological 80/20 train/test split is established first at the file level, followed by systematic sampling within each split to reduce the risk of leakage across the evaluation boundary. On the 746,094-record test set, the proposed hybrid IDS achieved 99.73% accuracy, 99.89% precision, 99.83% recall, 99.86% F1-score, and a false positive rate of 4.77%. The corresponding confusion matrix contains TN = 16,683, FP = 836, FN = 1205, and TP = 727,370, yielding 95.23% specificity and 97.53% balanced accuracy. Standalone Random Forest marginally outperformed the hybrid model in raw accuracy and false positive rate; therefore, the contribution of the proposed framework is centered on deployment-oriented anomaly monitoring, drift awareness, and generalization rather than absolute superiority in static classification metrics. A leave-one-attack-family-out experiment withholding MITM-ArpSpoofing from training showed that the hybrid model detected 85.26% of the unseen attack-family samples, compared with 85.18% for Random Forest alone and 7.05% for Isolation Forest alone. These findings provide initial evidence of generalization to one held-out attack family but should not be interpreted as proof of broad zero-day detection capability. The framework is therefore positioned as a competitive IDS that combines supervised detection with anomaly monitoring and concept drift awareness for deployment-oriented IoT security. Full article

Background: Accurate histological grading of gliomas, distinguishing low-grade (LGG) from high-grade (HGG) lesions, remains a critical determinant of treatment planning and patient prognosis. This study systematically investigates the optimal combination of structural MRI sequences, pretrained convolutional neural network (CNN) architectures, and supervised classifiers for automated glioma grade classification. Methods: Deep features were extracted from four MRI sequences (T1, contrast-enhanced T1 (T1Gd), T2, and FLAIR) from the BraTS 2023 dataset using five pretrained CNNs: VGG16, ResNet50, DenseNet121, EfficientNetB0, and InceptionV3. Features were refined via LASSO selection and classified using Random Forest, Support Vector Machine, XGBoost, Gradient Boosting, k-Nearest Neighbors (KNNs), and a shallow deep neural network. A patient-level 80/20 training–test partition was employed, with five-fold cross-validation used within the training set for feature selection and hyperparameter optimization. A total of 120 modality–extractor–classifier configurations were benchmarked on the held-out test set. Results: InceptionV3-derived features paired with KNN classifiers consistently yielded superior performance. Based on a TOPSIS multi-criteria ranking integrating accuracy, precision, recall, F1-score, and AUC, the best-performing configuration combined T1Gd features with KNN (closeness coefficient = 0.971; accuracy = 0.975, precision = 0.997, recall = 0.948, F1-score = 0.969, AUC = 0.996). The second-ranked configuration, using T1 features with KNN (closeness coefficient = 0.962), achieved significantly higher accuracy, recall, and F1-score, despite its slightly lower composite ranking. Conclusion: Deep feature extraction using InceptionV3 from T1-weighted MRI, coupled with KNN classification, represents a promising and practical approach for slice-level glioma grading and merits further validation in prospective clinical cohorts. Full article

Optical sensing arrays have emerged as transformative tools for disease diagnosis, offering low-cost, rapid, and multiplexed fingerprint detection capabilities. However, the high-dimensional and complex data generated by these arrays pose significant challenges for conventional analytical methods. The integration of machine learning (ML) has advanced this field by enabling automated feature extraction, robust pattern recognition, and accurate disease classification. This review provides a systematic overview of ML-reinforced optical sensing arrays, with a particular focus on three major modalities: colorimetric, fluorescent, and surface-enhanced Raman scattering (SERS) sensor arrays. We critically evaluate how ML algorithms—encompassing unsupervised, supervised, and deep learning paradigms—synergistically enhance the diagnostic performance of each sensor modality. Representative applications are highlighted, demonstrating high accuracy in detecting cancers and infectious diseases. Finally, we discuss the pressing challenges related to data standardization, model interpretability, and clinical translation, while outlining future directions toward intelligent, point-of-care, and personalized diagnostic systems. Full article

Background/Objectives: Cervical pedicle screw (CPS) placement provides superior biomechanical stability but remains technically demanding and associated with a risk of screw malposition. While recent advances in imaging and navigation have improved placement accuracy, reliable prediction of malposition remains challenging. The aim of this study was to evaluate whether machine learning (ML) models can predict CPS malposition using structured clinical and anatomical features. Methods: We performed a retrospective analysis of 862 pedicle screws from 168 posterior cervical spine surgeries conducted at our institution between 2018 and 2025. Clinical, procedural, and anatomical variables, including age, sex, body size parameters, surgical indication, vertebral level, pedicle angle, and pedicle width, were evaluated. Pedicle morphology was partially derived from CT-based automated segmentation using TotalSegmentator (v2.13.0), while selected anatomical parameters were manually measured. Supervised ML models, including Random Forest, Balanced Random Forest, XGBoost (v3.2.0), Support Vector Machine, and K-Nearest Neighbor, were trained and compared using Python and scikit-learn to predict inaccurate screw placement. Model performance was evaluated using Area Under the Receiver Operating Characteristic Curve (ROC AUC), F1-score, precision, and recall. Model interpretability was assessed using Shapley Additive Explanations (SHAP). Results: The dataset showed a clinically representative class distribution, with 91.1% of screws classified as acceptable and 8.9% as inaccurate. Across all models, predictive performance was moderate and consistent. Balanced Random Forest achieved the highest discriminative performance (ROC AUC 0.69) and provided the most balanced classification profile, while other models demonstrated comparable overall performance with varying sensitivity to the minority class. SHAP analysis identified anatomical and procedural variables, including pedicle width and angle, as relevant contributors to model output. Feature contributions were distributed across variables, with substantial overlap between outcome groups. Conclusions: ML-based prediction of CPS malposition using clinical and anatomical features demonstrates consistent and interpretable performance. The results highlight that predictive performance is primarily influenced by dataset characteristics, including class distribution and feature overlap, rather than model selection alone. This study provides an important baseline for ML-based CPS prediction and supports future research integrating larger datasets and more detailed anatomical representations to enhance predictive accuracy. Full article

This paper presents a hybrid computer vision framework that explicitly decouples geometric stem measurement from visual foliage condition classification for automated post-harvest quality assessment of Ruscus hypophyllum ornamental foliage. The proposed approach addresses a gap in the literature where heterogeneous quality attributes are typically treated within a single unified learning framework. In the first stage, stem size is estimated using a pixel-based geometric method that incorporates trigonometric orientation correction and spatial calibration via a reference marker of known length, enabling accurate conversion of image measurements to real-world physical dimensions. In the second stage, foliage condition is classified as good or poor using supervised machine learning models trained on Bag of Features representations extracted with the SIFT descriptor. A dataset of 1233 Ruscus hypophyllum images was acquired under controlled conditions using a consumer-grade smartphone camera and processed using open-source Python 3.11 libraries. Twenty-four classifier configurations across six model families were evaluated using stratified 10-fold cross-validation. The geometric estimation stage achieved a Mean Absolute Error (MAE) of 1.2 mm, a Root Mean Square Error (RMSE) of approximately 1.3 mm, and a size categorization accuracy of 99.84%. For foliage condition classification, the Linear Support Vector Machine achieved the best performance, with an accuracy of $92.4\pm 1.0$% and an F1-score of $91.1\pm 1.2$%, outperforming all other evaluated configurations. The proposed framework provides an interpretable, computationally efficient, and accessible solution for automated foliage quality grading, with potential applications in export-oriented ornamental foliage processing facilities. Full article

This study evaluates the potential of UV–Vis absorption and fluorescence detection (FLD), combined with chemometrics for the characterization of red wines and the preliminary screening of their Protected Designation of Origin (PDO) and grape variety. More than 100 samples were analyzed under both acidic and basic conditions using flow injection analysis (FIA), as it offers advantages in terms of automation, speed, and online coupling multichannel UV–Vis and FLD detectors, enabling the acquisition of complementary datasets. Spectral data generated were analyzed using exploratory and supervised chemometric methods. Results showed comparable performance across different data types. Although the discriminant information was limited, data exploration by Principal Component Analysis (PCA) revealed meaningful differences related to PDO and grape variety. Partial Least Squares Discriminant Analysis (PLS-DA) and Soft Independent Modeling of Class Analogies (SIMCA) provided remarkable classification results, with good overall performance, supporting the potential of this approach as a fast-screening tool for wine authentication and quality control, particularly in high-throughput analytical settings. Full article

Effective disassembly process planning is fundamental to the sustainability of power battery recycling. However, existing knowledge graph (KG) methods rely on flat ontologies, failing to capture multi-level semantic structures and depending heavily on manual annotation, which cannot meet the scalability demands of Industry 5.0. We propose TSO-Prompt, a triple-stage ontology prompt-driven method. First, a Battery–Component–Operation–Tool (BCOT) ontology model defines four entity types and four relationship types with strict domain and range constraints. Second, a three-stage prompt strategy is designed: Stage 1 (Pattern Injection) embeds BCOT definitions for simultaneous entity recognition, ontology classification, and relation extraction; Stage 2 (Temporal Completion) captures cross-step operational dependencies; Stage 3 (Ontology Self-Check) filters hallucination-induced errors via rule-based verification. The fully zero-shot pipeline requires no manual annotation. Experiments on 172 disassembly steps from five battery models show TSO-Prompt achieves 100% core semantic retention, 90.1% operation recognition accuracy, a 6.4% entity F1 improvement over supervised baselines, and 60% query path length reduction over flat graphs, validating its potential for automated KG construction aligned with Industry 5.0 objectives. Full article

At SAE Level 2 automation, the human driver retains full supervisory responsibility, making unobtrusive monitoring relevant for maintaining supervision under real-world driving conditions. Driver monitoring systems capable of operating robustly under such conditions are therefore essential, but wearable-based personalized approaches remain underexplored, particularly when the target labels are derived from experimental scenarios. This study presents a real-world SAE Level 2 on-road acquisition campaign and evaluates a target-driver intra-subject classification approach using non-intrusive wrist-derived signals. Physiological and motion data recorded with the Empatica E4 wristband, including blood volume pulse, electrodermal activity, heart rate, skin temperature, and triaxial wrist acceleration, were converted into image representations and processed with a frozen ResNet-50 feature extractor, principal component analysis, and a supervised classifier. The labels were scenario-derived operational driver-state classes defined from experimental phases and scenario groups. Personalization was assessed via a Leave-One-Experience-Out protocol on the target driver. Classification accuracy was 50% under external-user-only training, 54% under mixed target/external-user training, and 60% under target-driver-only training, with the target-driver-only configuration yielding the highest mean performance in the evaluated setting. For the low-demand baseline class, the one-vs.-rest classifier achieved 88.4% accuracy and an F1-score of 70%. These results provide initial evidence of the feasibility of personalized wrist-worn classification of scenario-derived operational driver-state classes under the real-world automated driving conditions evaluated in this study. Full article

Diabetic retinopathy (DR) grading requires reliable five-grade severity assessment under substantial acquisition variability and cross-dataset distribution shift. We propose PRISM-DR, a multi-objective five-grade DR grading framework trained under a gradient-partitioned strategy. The architecture is organized as a feedforward pipeline: a data-driven preprocessing stage followed by a ConvNeXtV2-Base backbone, a Recurrent BiFPN neck for multi-scale feature fusion, a Frequency-Aware Fusion module, a lightweight multi-scale reasoning transformer, dual classification heads with gradient-isolated pathways (categorical and ordinal), and a prototype memory module for embedding regularization. The CORAL ordinal head operates through a dedicated projection layer and is gradient-isolated from the backbone; the backbone is shaped by the cross-entropy, prototype contrastive, and view-consistency objectives, which carry indirect ordinal signal through severity-weighted class penalties and grade-indexed cluster regularization. The model is trained in a multi-crop setting with a phased loss curriculum designed for severely imbalanced DR datasets. Evaluated across six datasets under Fixed-Source, Multi-Target (FSMT) protocols, PRISM-DR trained on EyePACS + DDR achieves QWK of 0.835 on IDRiD, 0.865 on APTOS2019, and 0.720 on Messidor-2, with in-domain QWK = 0.920 and AUC-PR = 0.941 on EyePACS, outperforming RETFound, RETFound-Green, and MedGemma-4B in AUC-PR across all evaluated datasets. Quantitative interpretability evaluation against 755 expert-annotated lesion images yields 8.0× Energy Ratio Enrichment and a FAF gate retention ratio of 4.4× inside lesion regions, confirming that anatomically plausible spatial priors emerge from grade-level supervision alone, without pixel-level annotation. PRISM-DR establishes a superior accuracy–robustness–capacity trade-off for scalable, automated DR screening. Full article

This study investigates automated fake news detection as a reliability-oriented text classification problem in dynamic digital information environments. We propose H-FuseNet, a hybrid multi-representation fusion framework that combines pretrained transformer representations with deception-oriented handcrafted linguistic, stylistic, and semantic features. Using the WELFake dataset, we benchmark 15 baseline models, including classical classifiers, ensemble methods, recurrent and convolutional networks, and transformer fine-tuning models, under stratified 10-fold cross-validation with nested hyperparameter optimization. To examine generalization beyond a single benchmark, we train exclusively on WELFake and evaluate cross-dataset performance on three held-out external datasets: FakeNewsNet, CoAID, and LLM-generated misinformation. H-FuseNet integrates transformer document embeddings with a lightweight feature-processing MLP, optional contextual feature streams when metadata are available, and auxiliary supervision through pseudo-labeled headline body stance and clickbait signals. The proposed model achieves 98.9% mean accuracy and 0.998 ROC–AUC, while maintaining strong calibration, with a Brier score of 0.012 and Expected Calibration Error of 0.009, and low variance across folds. Cross-dataset evaluation yields accuracies of 87.34% on FakeNewsNet, 83.56% on CoAID, and 91.22% on LLM-generated misinformation, demonstrating robust generalization under distribution shift. Ablation analyses show that handcrafted features, auxiliary tasks, and learned fusion each contribute to performance, while Wilcoxon and McNemar tests indicate statistically significant differences against selected strong baselines. Error analysis shows that remaining failures mainly occur in professionally written misinformation that imitates neutral journalistic style. Overall, the results suggest that calibrated multi-representation fusion can improve the reliability of automated fake news detection systems. Full article

Automated classification of Arabic legal texts presents unique challenges stemming from the formal register of judicial language, domain-specific Sharī‘a terminology, and the severe class imbalance inherent in hierarchical legal taxonomies. This paper addresses these challenges through a systematic investigation of hierarchical multi-class classification applied to a dataset of 1146 Arabic judicial cases sourced from the Saudi Ministry of Justice open data portal. Cases are annotated at two hierarchical levels: a binary main-class distinction between criminal and civil cases ($L_1$), and a fine-grained sub-category classification across 63 legal topics ($L_2$), exhibiting a class imbalance ratio of 43:1. We evaluate three methodological families under four classification architectures. For encoder-only models, we fine-tune AraBERT, CAMeL-BERT, MARBERTv2, and XLM-RoBERTa under Flat (Bottom-Up), Hierarchical Multi-Task, Two-Stage, and Hierarchy-Aware configurations. For encoder–decoder models, we adapt AraT5v2 under the same four architectures. Finally, we conduct a prompt-based evaluation of GPT-4o under zero-shot and targeted few-shot settings. The experimental results demonstrate that the Hierarchy-Aware architecture consistently achieves the strongest fine-grained classification performance across both model families. AraBERTv2 with Hierarchy-Aware training achieves the best overall $L_2$ accuracy of 83.04% and a Macro-F1 of 76.07%, while the Single Multi-Task configuration achieves the highest $L_1$ accuracy of 99.57%. GPT-4o under 5-shot prompting achieves 99.13% $L_1$ accuracy and a competitive $L_2$ Macro-F1 of 74.30% without task-specific fine-tuning, though supervised models maintain stronger overall fine-grained performance. These findings highlight the importance of explicit hierarchical supervision and domain-adapted pre-training for Arabic legal text classification, and establish strong baselines for future research in this underexplored domain. Full article

The Brazilian Cerrado, a global biodiversity hotspot, is under increasing pressure from agricultural expansion and native vegetation conversion, underscoring the need for efficient monitoring to support conservation and environmental policies. In heterogeneous landscapes, land use and land cover (LULC) mapping using supervised classification methods faces a major bottleneck: the need for extensive and high-quality training datasets. To address this challenge, we propose a semi-automated, clustering-based methodology for mapping secondary vegetation within previously deforested areas, reducing training-sample requirements and enabling scalable mapping through the clustering of satellite image time series. In the first stage, an unsupervised process integrates graphics processing unit (GPU)-accelerated Self-Organizing Maps and hierarchical clustering with Dynamic Time Warping to produce spectro-temporal clusters. In the second stage, specialists label and refine these clusters by visual interpretation, transferring expert knowledge from individual pixels to grouped spectro-temporal patterns. Applied to 692,000 km² of previously deforested land in the Cerrado biome, the methodology produced a mapped secondary vegetation area of 81,209 km² (11.74%). The design-based estimated area was 98,683 ± 10,071 km², with an overall accuracy of 96.45 ± 1.52%, a user’s accuracy of 96.27 ± 2.40%, a producer’s accuracy of 79.22 ± 7.94%, and an F1-score of 86.90%. The initial cluster labeling accounted for 86.3% of the final secondary vegetation area and limited the interpretation task to approximately 3000 cluster-level decisions. Implemented in the TerraClass Cerrado 2024 cycle, the workflow reduced the secondary vegetation mapping phase from approximately two years to six months while maintaining the thematic accuracy required for large-scale operational monitoring. Full article

The exponential growth of Indicators of Compromise (IoCs) has overwhelmed manual triage processes in Security Operations Centers (SOCs), necessitating automated solutions for large-scale log analysis. This study proposes a hybrid machine learning framework that integrates supervised and unsupervised learning to automate the classification, clustering, and contextual interpretation of Cyber Threat Intelligence (CTI). The primary contribution lies in a multi-stage feature engineering pipeline that enriches raw SIEM logs with cyclical temporal encoding and geographical metadata. In the supervised phase, a comparative evaluation of gradient boosting classifiers—XGBoost, LightGBM, and CatBoost—demonstrates that all three achieve competitive performance in categorizing known attack techniques, consistently outperforming the Random Forest baseline. The results indicate that classifier performance is dataset-dependent, and practitioners are encouraged to select the most suitable model based on their operational environment. Simultaneously, the unsupervised phase employs density-based clustering to identify emerging and previously unknown threat patterns by correlating adversarial behaviors with source attribution. By combining these two approaches, the framework ensures near-real-time feasibility and significantly enhances the scalability of automated threat extraction from distributed honeypot environments. Full article