Search Results (359)

Automated ocular disease diagnosis faces critical challenges including insufficient diagnostic precision, local–global feature imbalance, rigid feature fusion, weak cross-domain generalization, and difficult lightweight deployment. This study aims to develop a high-performance, generalizable, and deployable hybrid deep learning architecture for accurate multi-class ocular disease diagnosis. We propose the Res101-MViT-Ens hybrid architecture, which fuses ResNet101 for local fine-grained feature extraction and MobileViT-XXS for global contextual modeling via an end-to-end dynamic learnable weight fusion mechanism, with class-balanced sampling and medically adaptive augmentation for data preprocessing. The model is validated on the ODIR-5K dataset and cross-evaluated on three heterogeneous datasets (MESSIDOR-2, Kaggle DR, EyePACS). It achieves 99.44% accuracy, a 99.41% F1-score, and 99.32% Kappa on ODIR-5K, with a 99.46% average cross-dataset accuracy, outperforming state-of-the-art models. With 54 M parameters and 42.6 ms per-image inference latency on the Snapdragon 8 Gen2 edge module (Qualcomm Technologies, Inc., San Diego, CA, USA), it outperforms mainstream edge architectures. This proposed architecture achieves state-of-the-art diagnostic precision; balances accuracy, generalization and practicality; and is suitable for lightweight grassroots deployment in ocular disease screening. Full article

Infected URLs are always regarded as a serious threat to cybersecurity, serving as pathways to phishing, maliciousness, and other offenses. Although transformer-based models have demonstrated good performance in malicious URL detection, their high computational cost and latency make them impractical for deployment in real-time or resource-constrained systems. Allocated on the basis of knowledge distillation (KD), lightweight models tend to be efficient but are commonly not sufficiently discriminative to distinguish between malicious and benign URLs with non-cataclysmic lexical overlaps, particularly when dealing with an imbalanced dataset. In order to address these issues, we propose Contra-KD, a lightweight transformer model that incorporates contrastive learning (CL) and KD. This proposed framework imposes structured embedding matching, allowing the student model to learn more meaningful and generalized depictions. Contra-KD uses a compact 6-layer student transformer architecture based on ELECTRA to scale parameters up and can achieve more than 90% computational fidelity with a high accuracy. In this scheme, CL improves the feature of discrimination by semantically clustering similar URLs and separating different URLs. This tendency serves to limit confusion, especially when a common lexical trait is held between two words and/or in the presence of adversarial obfuscation. Through a large-scale publicly available Kaggle dataset of 651,191 URLs in imbalanced scenarios, the proposed Contra-KD can achieve 99.05% accuracy, 99.96% ROC-AUC, and 98.18% MCC which are superior to their counterparts including lightweight models and transformer-based ones. To summarize, Contra-KD proposes an efficient transformer architecture that is both small and effective in computation while delivering stable detection performance. Full article

Background: Artificial intelligence (AI) shows promising results in lymphoma detection, prediction, and classification. However, translating these findings into practice requires a rigorous assessment of potential biases, clinical utility, and further validation of research models. Objective: The goal of this study was to summarize existing studies on artificial intelligence models for the histopathological detection of lymphoma. Design: This study adhered to the PRISMA Extension for Scoping Reviews (PRISMA-ScR) guidelines. A systematic search was conducted across three major databases (Scopus, PubMed, Web of Science) for English-language articles and reviews published between 2016 and 2025. Seven precise search queries were applied to identify relevant publications, accounting for variations in study modality, algorithmic architectures, and disease-specific terminology. Results: The search identified 612 records, of which 36 articles met the inclusion criteria. These studies presented 36 AI models, comprising 30 diagnostic and six prognostic applications, with Convolutional Neural Networks (CNNs) being the predominant architecture. Regarding data sources, 83% (30/36) of datasets utilized Hematoxylin and Eosin (H&E)-stained images, while the remainder relied on diverse modalities, including IHC-stained slides, bone marrow smears, and other tissue preparations. Studies predominantly utilized retrospective, private cohorts with sample sizes typically ranging from 50 to 400 patients; only a minority leveraged open-access repositories (e.g., Kaggle, TCGA). The primary application was slide-level multi-class classification, distinguishing between specific lymphoma subtypes and non-neoplastic controls. Beyond diagnosis, a subset of studies explored advanced prognostic tasks, such as predicting chemotherapy response and disease progression (e.g., in CLL), as well as automated biomarker quantification (c-MYC, BCL2, PD-L1). Reported diagnostic performance was generally high, with accuracy ranging from 60% to 100% (clustering around 90%) and AUC values spanning 0.70 to 0.99 (predominantly >0.90). Conclusions: While AI models demonstrate high diagnostic accuracy, their translation into practice is limited by unstandardized protocols, morphological complexity, and the “black box” nature of algorithms. Critical issues regarding data provenance, image noise, and lack of representativeness raise risks of systematic bias, hence the need for rigorous validation in diverse clinical environments. Full article

Organizational productivity and workforce management are highly affected by employee attrition. Thus, an employee attrition prediction system may allow human resource management to enhance the workplace by minimizing attrition. This study proposes a new and interpretable ensemble learning framework for employee attrition prediction. The model integrates SHapley Additive exPlanations (SHAP)-based feature selection, Optuna hyperparameter optimization, and dual explainability using SHAP and Local Interpretable Model-agnostic Explanations (LIME). Random oversampling (ROS) is used to address class imbalance. The proposed framework allows for both global and local interpretability, enabling actionable insights into retention drivers. It was assessed using two benchmark datasets: the Kaggle HR Analytics dataset (14,999 records) and the IBM HR dataset (1470 records). The results revealed that the most impactful factors on employee attrition are promotion history, tenure, job satisfaction, workload, average monthly hours, overtime, and financial incentives. Furthermore, the proposed model achieved exceptional performance on both datasets. On the Kaggle dataset, it reached an accuracy of 98.72%, an F1-score of 97.29%, and an ROC–AUC of 0.994, while on the IBM dataset, it produced an accuracy of 97.72%, an F1-score of 97.74%, and an ROC–AUC of 0.995. Moreover, the proposed approach shows high computational efficiency, demonstrating that it is suitable for real-world deployment. These findings indicate that integrating explainable AI techniques, resampling tools, and automated hyperparameter tuning can achieve robust, accurate, and actionable employee attrition predictions, supporting HR managers’ decision-making. Full article

Manual microscopy for malaria diagnosis is labor-intensive and prone to inter-observer variability. This study presents an automated binary classification approach for detecting malaria ring-form infections in thin blood smear single-cell images using a parallel neural network framework. Utilizing a balanced Kaggle dataset of 27,558 erythrocyte crops, images were standardized to 128 × 128 pixels and subjected to on-the-fly augmentation. The proposed architecture employs a dual-branch fusion strategy, integrating a convolutional neural network for local morphological feature extraction with a multi-head self-attention branch to capture global spatial relationships. Performance was rigorously evaluated using 10-fold stratified cross-validation and an independent 10% hold-out test set. Results demonstrated high-level discrimination, with all models achieving an ROC–AUC of approximately 0.99. The primary model (Model#1) attained a peak mean accuracy of 0.9567 during cross-validation and 0.97 accuracy (macro F1-score: 0.97) on the independent test set. In contrast, increasing architectural complexity in Model#3 led to a performance decline (0.95 accuracy) due to higher false-positive rates. These findings suggest that moderate-capacity feature fusion, combining convolutional descriptors with attention-based aggregation, provides a robust and generalizable solution for automated malaria screening without the risks associated with over-parameterization. Despite a strong performance, immediate clinical use remains limited because the model was developed on pre-segmented single-cell images, and external validation is still required before routine implementation. Full article

Convolutional neural networks (CNNs) can achieve high accuracy in photovoltaic (PV) fault recognition from RGB imagery, yet their decisions may rely on shortcut cues induced by heterogeneous backgrounds, viewpoints, and class imbalance. This work presents a multi-method explainability audit on the Kaggle PV Panel Defect Dataset (six classes), comparing five architectures (Baseline CNN, VGG16, ResNet50, InceptionV3, EfficientNetB0). Explanations are obtained with LIME superpixel surrogates (reported together with kernel-weighted surrogate fidelity), occlusion sensitivity (quantified via IoU@Top10% against consistent proxy masks, Shannon entropy, and Hoyer sparsity), and Integrated Gradients evaluated by deletion–insertion faithfulness and a Faithfulness Gap. While ResNet50 yields the best predictive performance, EfficientNetB0 shows the most consistent faithfulness evidence and stable panel-centered attributions. The analysis highlights class-dependent vulnerability to context cues, especially for the Clean and damaged classes, and supports using quantitative explainability diagnostics during model selection and dataset curation to mitigate shortcuts in vision-based PV monitoring. Full article

Short-term photovoltaic (PV) power forecasting is crucial for secure and economical grid operation, yet remains challenging under fast and nonstationary irradiance fluctuations. This paper presents a plant-level TA–SH–LSTM framework that integrates temporal attention into an LSTM encoder to highlight informative subsegments for improved ramp tracking and peak localization and applies budget-aware Successive Halving to jointly tune window length and key hyperparameters under a fixed training budget. To enhance PV-engineering interpretability, we establish a first-order thermal inertia surrogate that explicitly links module temperature to ambient temperature and irradiance, and evaluate robustness across irradiance-tercile regimes within the observation window. Experiments on two real PV plants from the Kaggle Solar Power Generation dataset demonstrate consistent gains over a baseline LSTM and an SH-tuned LSTM. On Plant 1, MAE/RMSE decreases from 1141.1/2066.6 kW to 223.4/424.6 kW and $R^2$ increases from 0.932 to 0.997. Without retraining, the model transfers to Plant 2 with 286.1 kW MAE, 477.1 kW RMSE, and $R^2$ = 0.993, confirming strong cross-site generalization and practical utility under varying operating conditions. Full article

Infectious diseases are contributing to a major public health challenge worldwide, affecting individuals across all age groups and regions. An infectious disease is a pathological condition caused by harmful microorganisms. These are bacteria, viruses, fungi, or parasites that enter the body, multiply, and disturb normal physiological functions, leading to clinical manifestations. At present, the detection of infectious disease is mainly based on vital signs and a limited set of biomarkers. This limited approach fails to fully capture the complications of infection-related physiological changes. To address these limitations, vital signs and a broad range of hematological and biochemical biomarkers are integrated with machine learning and explainable artificial intelligence (XAI). The data set used in this study was collected from the Kaggle data source. The dataset consists of vital sign values, such as body temperature, systolic and diastolic blood pressure, respiratory rate, heart rate, and oxygen saturation, along with blood-based biomarkers including albumin, base excess, bicarbonate, bilirubin, blast cells, calcium, creatinine, gamma-glutamyl transferase (GGT), glucose, hematocrit, hemoglobin, lactate, leukocytes, neutrophils, C-reactive protein (CRP), platelets, potassium, sodium, alanine aminotransferase (TGP/ALT), activated partial thromboplastin time (TTPA), and urea. These parameters provide a complete view of the patient’s physiological and biochemical state during infection. Feature selection was performed using a hybrid approach combining correlation filtering, mutual information, tree-based feature importance, and XAI validation (SHAP, permutation sensitivity) to ensure both predictive accuracy and interpretability. The integration of these techniques supports accurate classification and AI-assisted decision-making. The findings of this study highlight the importance of integrating both vital sign monitoring and laboratory assessments for effective infectious disease management. Full article

Industry 5.0 promotes sustainable, resilient, and human-centric manufacturing. Many factories struggle to produce energy-aware schedules that balance throughput, energy, and time-of-use (TOU) tariffs. Classical methods (heuristics and optimization) help but lack transparency and adaptability, limiting operator-in-the-loop use. Generative AI, particularly Large Language Models (LLMs), offers reasoning, adaptation, and interaction, yet integration with production scheduling is nascent. We introduce a hybrid framework that combines classical heuristics with GPT-4 reasoning to create an Industry 5.0-compatible Co-Pilot for energy-aware factory scheduling. The Co-Pilot evaluates and adapts machine operation schedules to avoid peak windows and explains trade-offs in natural language. We evaluate on three datasets (CTU synthetic, Kaggle manufacturing, Zenodo benchmark). Results show the heuristic Co-Pilot consistently reduces peak load share versus classical baselines at similar cost; on Zenodo, GPT-4 saves 4–7% in cost and energy, while its performance is less stable on synthetic data. These findings highlight the promise of LLM-powered scheduling and the value of hybrid human-AI strategies in Industry 5.0. Full article

Artificial Intelligence is advancing rapidly, raising critical concerns about the integrity of digital content, particularly in sensitive domains such as medical imaging. Recent AI techniques, such as Generative Adversarial Networks (GANs) and diffusion models, can generate highly realistic synthetic medical images, posing risks of misdiagnosis, inappropriate treatment, and other adverse outcomes. This paper presents a deep learning-based approach to distinguish between authentic and synthetic images of skin malignancies generated by DCGAN, Wasserstein GAN (WGAN), and Stable Diffusion. A comprehensive dataset was constructed using authentic malignant skin images from an open-source Kaggle repository, alongside artificially generated images. Multiple deep learning models were trained and evaluated, with DenseNet169 achieving the highest performance, reaching 99.67% training accuracy, 97.50% validation accuracy, and 98.50% test accuracy—along with substantial precision, recall, and F1 scores across all classes. These results demonstrate the model’s efficacy in identifying both real and fake medical images. This work contributes to the emerging field of medical image forensics, highlighting its potential integration into clinical and insurance workflows to prevent fraud, strengthen trust, and mitigate risks. Furthermore, it lays the groundwork for future studies involving larger datasets, additional Deepfake generation methods, and real-time clinical applications. Full article

Ground-based and space-based gravitational wave (GW) detectors cover complementary frequency bands, laying the foundation for future multi-band collaborative observations. Detecting weak signals within non-stationary noise remains challenging. To address this, we propose a Physics-Guided Dynamic Sparse Attention (PGDSA) framework. The framework introduces a differentiable wavelet layer to explicitly embed sensitive frequency bands and time–frequency priors while utilizing intra-block Top-K sparse attention for efficient long-range temporal modeling. Training is performed on space-based simulation data with joint optimization for signal detection and waveform reconstruction. We then evaluate detection performance and zero-shot transfer capability on ground-based data. Experimental results show that PGDSA achieves an ROC-AUC of 0.886 on the Kaggle G2Net private leaderboard. On GWOSC O3 real data, the model yields high confidence scores for confirmed binary black hole events. On LISA simulation data, the framework achieves detection rates exceeding 99% for multiple signal types (SNR = 50, FAR = 1%) with waveform reconstruction Overlap comparable to baseline methods. These results demonstrate that PGDSA enables unified modeling across both space-based and ground-based scenarios. Full article

Autism Spectrum Disorder (ASD) is a neurological condition that affects communication and social interaction skills, with individuals experiencing a range of challenges that often require specialized care. Automated systems for recognizing ASD face significant challenges due to the complexity of identifying distinguishing features from facial images. This study proposes an incremental advancement in ASD recognition by introducing a dual-stream model that combines handcrafted facial-landmark features with deep learning-based pixel-level features. The model processes images through two distinct streams to capture complementary aspects of facial information. In the first stream, facial landmarks are extracted using MediaPipe (v0.10.21),with a focus on 137 symmetric landmarks. The face’s position is adjusted using in-plane rotation based on eye-corner angles, and geometric features along with 52 blendshape features are processed through Dense layers. In the second stream, RGB image features are extracted using pre-trained CNNs (e.g., ResNet50V2, DenseNet121, InceptionV3) enhanced with Squeeze-and-Excitation (SE) blocks, followed by feature refinement through Global Average Pooling (GAP) and DenseNet layers. The outputs from both streams are fused using weighted concatenation through a softmax gate, followed by further feature refinement for classification. This hybrid approach significantly improves the ability to distinguish between ASD and non-ASD faces, demonstrating the benefits of combining geometric and pixel-based features. The model achieved an accuracy of 96.43% on the Kaggle dataset and 97.83% on the YTUIA dataset. Statistical hypothesis testing further confirms that the proposed approach provides a statistically meaningful advantage over strong baselines, particularly in terms of classification correctness and robustness across datasets. While these results are promising, they show incremental improvements over existing methods, and future work will focus on optimizing performance to exceed current benchmarks. Full article

The rapid spread of diseases like COVID-19 highlights the need for adaptable monitoring systems to support public health measures such as mask compliance and social distancing. This study presents the CB-OWL-ViT framework: a Cluster-Based Open-World Localization Vision Transformer for mask detection and social distance estimation. It incorporates homography-based distance estimation for effective deployment with monocular cameras. The innovative integration of open-world vision-language detection with a clustering-based strategy enhances mask-wearing assessments, enabling adaptability without retraining. Evaluations on datasets including Kaggle, Roboflow, and a new dataset from the University of Waterloo show that CB-OWL-ViT improves mask detection precision by 0.37 and F1-score by 0.2 compared to the baseline. The homography module achieves a Mean Absolute Error of 0.1116 in distance estimation, and real-world tests demonstrate a recall of 0.98 for detecting noncompliance in the “Without Mask” class. This framework is a practical solution for large-scale disease monitoring across various settings. Full article

Computer-aided diagnosis (CAD) systems based on deep learning have shown significant potential for Alzheimer’s disease (AD) stage classification from Magnetic Resonance Imaging (MRI). Nevertheless, challenges such as class imbalance, small sample sizes, and the presence of multiple slices per subject may lead to biased evaluation and statistically unreliable performance, particularly for minority classes. In this study, a Vision Transformer (ViT)-based framework is proposed for multi-class AD classification using a Kaggle dataset containing 6400 MRI slices across four cognitive stages. A subject-wise data-splitting strategy is employed to prevent information leakage between the training and testing sets, and the statistical unreliability of near-perfect scores in underrepresented classes is critically examined. An ablation study is conducted to assess the contribution of key architectural components, demonstrating the effectiveness of self-attention and patch embedding in capturing discriminative features. Furthermore, attention-based visualization maps are incorporated to highlight brain regions influencing the model’s decisions and to illustrate subtle anatomical differences between MildDemented and VeryMildDemented cases. The proposed approach achieves a test accuracy of 97.98%, outperforming existing methods on the same dataset while providing improved interpretability. It supports early and accurate AD stage identification. Full article

Short-term electricity load forecasting has become increasingly challenging due to growing demand volatility, nonlinear load patterns, and the dynamic penetration of renewable energy sources. Conventional forecasting models often suffer from sensitivity to hyperparameter settings and limited capability in capturing long-term temporal dependencies. To address these issues, this paper proposes a hybrid forecasting framework that integrates an Improved Snow Ablation Optimizer (ISAO) with a Dilated Bidirectional Gated Recurrent Unit (Dilated BiGRU). The proposed ISAO enhances the original Snow Ablation Optimizer through three key strategies to improve performance in high-dimensional optimization problems: (i) a subgroup cooperative mechanism to alleviate cross-dimensional interference, (ii) a learning-automata-based adaptive dimension assignment strategy to dynamically allocate optimization resources, and (iii) a t-distribution-based adaptive step size mechanism to balance global exploration and local exploitation. Extensive experiments on the CEC2017 benchmark suite demonstrate that ISAO achieves superior convergence speed and optimization accuracy, with average rankings of 1.60, 1.77, and 2.03 on 30-, 50-, and 100-dimensional problems, respectively, significantly outperforming the original SAO and several state-of-the-art metaheuristic algorithms. Building upon this optimization capability, ISAO is employed to automatically tune the key hyperparameters of the Dilated BiGRU model. Experiments conducted on the Kaggle electricity load dataset show that the proposed ISAO-Dilated BiGRU model achieves MAE, MAPE, and RMSE values of 20.003, 1.711%, and 25.926, respectively, corresponding to reductions of 16.6%, 15.6%, and 17.7% compared with the baseline model, along with an R² of 0.97841. Comparative results against RNN, LSTM, Random Forest, and the original Dilated BiGRU confirm the robustness and superior long-term dependency modeling capability of the proposed framework. Overall, the proposed ISAO effectively enhances hyperparameter optimization quality and significantly improves the predictive accuracy and stability of the Dilated BiGRU model, providing a reliable and practical solution for short-term electricity load forecasting in modern power systems. Full article