Search Results (378)

We propose Federated Entropic High-Order Descent (FedEHD), a drop-in client optimizer that augments local SGD with (i) an entropy (sign) term and (ii) quadratic and cubic gradient components for drift control and implicit clipping. Across non-IID CIFAR-10 and CIFAR-100 benchmarks (100 clients, 10% sampled per round), FedEHD achieves faster and higher convergence than strong baselines including FedAvg, FedProx, SCAFFOLD, FedDyn, MOON, and FedAdam. On CIFAR-10, it reaches 70% accuracy in approximately 80 rounds (versus 100 for MOON and 130 for SCAFFOLD) and attains a final accuracy of 72.5%. On CIFAR-100, FedEHD surpasses 60% accuracy by about 150 rounds (compared with 250 for MOON and 300 for SCAFFOLD) and achieves a final accuracy of 68.0%. In an environmental monitoring case study involving four distributed air-quality stations, FedEHD yields the highest macro AUC/F1 and improved calibration (ECE 0.183 versus 0.186–0.210 for competing federated methods) without additional communication and with only $\mathcal{O}(d)$ local overhead. The method further provides scale-invariant coefficients with optional automatic adaptation, theoretical guarantees for surrogate descent and drift reduction, and convergence curves that illustrate smooth and stable learning dynamics. Full article

Circular economy learning fosters competencies in sustainable resource management and environmental protection, which have been recognized by the OECD (Organization for Economic Cooperation and Development) to be essential for cross-curricular training and higher education. However, implementing gamification techniques through mobile applications remains challenging, as their effectiveness depends on students’ willingness to adopt them. This study evaluated acceptance of a gamified mobile application for circular economy learning among university students in Peru, analyzing the relationships between the constructs of the Technology Acceptance Model (TAM). A quantitative correlational case study involving 76 students was conducted. The results showed a moderate-to-high acceptance rate of 73.69%, with significant correlations identified between the TAM constructs. This study contributes to closing gaps in empirical evidence on the acceptance of technology for sustainability education in diverse contexts. Future studies should integrate generative artificial intelligence into gamified apps to deliver personalized feedback and employ learning analytics tools for progress tracking, supporting global efforts toward SGD 4 (Quality Education) and SDG 12 (Responsible Production and Consumption) for the transition to circular economies. Full article

Objectives: This study aimed to develop and validate a biparametric MRI (bpMRI)-based radiomics model for the noninvasive prediction of tumor–stroma ratio (TSR) in prostate cancer (PCa). Additionally, we sought to explore lesion distribution patterns in the peripheral zone (PZ) and transition zone (TZ) to provide deeper insights into the biological behavior of PCa. Methods: This multicenter retrospective study included 223 pathologically confirmed PCa patients, with 146 for training and 39 for internal validation at Center 1, and 38 for external testing at Center 2. All patients underwent preoperative bpMRI (T2WI, DWI acquired with a b-value of 1400 s/mm², and ADC maps), with TSR histopathologically quantified. Regions of interest (ROIs) were manually segmented on bpMRI images using ITK-SNAP software (version 4.0.1), followed by high-throughput radiomic feature extraction. Redundant features were eliminated via Spearman correlation analysis and least absolute shrinkage and selection operator (LASSO) regression. Five machine learning (ML) classifiers—Logistic Regression (LR), Support Vector Machine (SVM), BernoulliNBBayes, Ridge, and Stochastic Gradient Descent (SGD)—were trained and optimized. Model performance was rigorously evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA). Results: The Ridge demonstrated superior diagnostic performance, achieving AUCs of 0.846, 0.789, and 0.745 in the training, validation, and test cohorts, respectively. Lesion distribution analysis revealed no significant differences between High-TSR and Low-TSR groups (p = 0.867), suggesting that TSR may not be strongly associated with zonal localization. Conclusions: This exploratory study suggests that a bpMRI-based radiomic model holds promise for noninvasive TSR estimation in prostate cancer and may provide complementary insights into tumor aggressiveness beyond conventional pathology. Full article

The gravity forward modeling algorithm is a compute-intensive method and is widely used in scientific computing, particularly in geophysics, to predict the impact of subsurface structures on surface gravity fields. Traditional implementations rely on CPUs, where performance gains are mainly achieved through algorithmic optimization. With the rise of domain-specific architectures, FPGA offers a promising platform for acceleration, but faces challenges such as limited programmability and the high cost of nonlinear function implementation. This work proposes an FPGA-based co-processor to accelerate gravity forward modeling. A RISC-V core is integrated with a custom instruction set targeting key computation steps. Tasks are dynamically scheduled and executed on eight fully pipeline processing units, achieving high parallelism while retaining programmability. To address nonlinear operations, we introduce a piecewise linear approximation method optimized via stochastic gradient descent (SGD), significantly reducing resource usage and latency. The design is implemented on the AMD UltraScale+ ZCU102 FPGA (Advanced Micro Devices, Inc. (AMD), Santa Clara, CA, USA) and evaluated across several forward modeling scenarios. At 250 MHz, the system achieves up to 179× speedup over an Intel Xeon 5218R CPU (Intel Corporation, Santa Clara, CA, USA) and improves energy efficiency by 2040×. To the best of our knowledge, this is the first FPGA-based gravity forward modeling accelerate design. Full article

Lung cancer, the leading cause of cancer-related mortality worldwide, necessitates better methods for earlier and more accurate detection. To this end, this study introduces LCxNet, a novel, custom-designed convolutional neural network (CNN) framework for computer-aided diagnosis (CAD) of lung cancer. The IQ-OTH/NCCD lung CT dataset, which includes three different classes—benign, malignant, and normal—is used to train and assess the model. The framework is implemented using five optimizers, SGD, RMSProp, Adam, AdamW, and NAdam, to compare the learning behavior and performance stability. To bridge the gap between model complexity and clinical utility, we integrated Explainable AI (XAI) methods, specifically Grad-CAM for decision visualization and t-SNE for feature space analysis. With accuracy, specificity, and AUC values of 99.39%, 99.45%, and 100%, respectively, the results demonstrate that the LCxNet model outperformed the state-of-the-art models in terms of diagnostic performance. In conclusion, this study emphasizes how crucial XAI is to creating trustworthy and efficient clinical tools for the early detection of lung cancer. Full article

Optimizing concrete production requires balancing ingredient ratios and using local resources to produce an economical material with the desired consistency, strength, and durability. Compressive strength is crucial for structural design, yet predicting it accurately is challenging due to the complex interplay of various factors, including component types, water–cement ratio, and curing time. This study employs a Multi-layer Perceptron Neural Network (ANN_MLP) to model the relationship between input variables and the compressive strength of normal and high-performance concrete. A dataset of 1030 samples from the literature was used for training and evaluation. The optimized ANN_MLP configuration included 16 neurons in a single hidden layer, with the ‘tanh’ activation function and ‘sgd’ solver. It achieved an R² of 0.892, an MAE of 3.648 MPa, and an RMSE of 5.13 MPa. The model was optimized using a univariate sensitivity analysis to measure the impact of each hyperparameter on performance and select optimal values to maximize the accuracy and robustness. Full article

With global climate change, urbanization, and agricultural resource limitations, precision agriculture and crop monitoring are crucial worldwide. Integrating multi-source remote sensing data with deep learning enables accurate crop mapping, but selecting optimal network architectures remains challenging. To improve remote sensing-based fruit planting classification and support orchard management and rural revitalization, this study explored feature selection and network optimization. We proposed an improved CF-EfficientNet model (incorporating FGMF and CGAR modules) for fruit planting classification. Multi-source remote sensing data (Sentinel-1, Sentinel-2, and SRTM) were used to extract spectral, vegetation, polarization, terrain, and texture features, thereby constructing a high-dimensional feature space. Feature selection identified 13 highly discriminative bands, forming an optimal dataset, namely the preferred bands (PBs). At the same time, two classification datasets—multi-spectral bands (MS) and preferred bands (PBs)—were constructed, and five typical deep learning models were introduced to compare performance: (1) EfficientNetB0, (2) AlexNet, (3) VGG16, (4) ResNet18, (5) RepVGG. The experimental results showed that the EfficientNetB0 model based on the preferred band performed best in terms of overall accuracy (87.1%) and Kappa coefficient (0.677). Furthermore, a Fine-Grained Multi-scale Fusion (FGMF) and a Condition-Guided Attention Refinement (CGAR) were incorporated into EfficientNetB0, and the traditional SGD optimizer was replaced with Adam to construct the CF-EfficientNet architecture. The results indicated that the improved CF-EfficientNet model achieved high performance in crop classification, with an overall accuracy of 92.6% and a Kappa coefficient of 0.830. These represent improvements of 5.5 percentage points and 0.153, compared with the baseline model, demonstrating superiority in both classification accuracy and stability. Full article

Skid resistance is a key factor in road safety, directly affecting vehicle stability and braking efficiency. To enhance predictive accuracy, this study develops a multilayer perceptron (MLP) model for forecasting the Sideway Force Coefficient (SFC) of asphalt pavements and systematically examines the role of activation functions and optimizers. Seven activation functions (Sigmoid, Tanh, ReLU, Leaky ReLU, ELU, Mish, Swish) and three optimizers (SGD, RMSprop, Adam) are evaluated using regression metrics (MSE, RMSE, MAE, R²) and loss-curve analysis. Results show that ReLU and Mish provide notable improvements over Sigmoid, with ReLU increasing goodness of fit and accuracy by 13–15%, and Mish further enhancing nonlinear modeling by 12–14%. For optimizers, Adam achieves approximately 18% better performance than SGD, offering faster convergence, higher accuracy, and stronger stability, while RMSprop shows moderate performance. The findings suggest that combining ReLU or Mish with Adam yields highly precise and robust predictions under multi-source heterogeneous inputs. This study offers a reliable methodological reference for intelligent pavement condition monitoring and supports safety management in highway transportation systems. Full article

We aimed to investigate the utility of peritumoral edema-derived radiomic features from magnetic resonance imaging (MRI) image weights and fused MRI sequences for enhancing the performance of machine learning-based glioma grading. The present study utilized the Multimodal Brain Tumor Segmentation Challenge 2023 (BraTS 2023) dataset. Laplacian Re-decomposition (LRD) was employed to fuse multimodal MRI sequences. The fused image quality was evaluated using the Entropy, standard deviation (STD), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) metrics. A comprehensive set of radiomic features was subsequently extracted from peritumoral edema regions using PyRadiomics. The Boruta algorithm was applied for feature selection, and an optimized classification pipeline was developed using the Tree-based Pipeline Optimization Tool (TPOT). Model performance for glioma grade classification was evaluated based on accuracy, precision, recall, F1-score, and area under the curve (AUC) parameters. Analysis of fused image quality metrics confirmed that the LRD method produces high-quality fused images. From 851 radiomic features extracted from peritumoral edema regions, the Boruta algorithm selected different sets of informative features in both standard MRI and fused images. Subsequent TPOT automated machine learning optimization analysis identified a fine-tuned Stochastic Gradient Descent (SGD) classifier, trained on features from T₁Gd+FLAIR fused images, as the top-performing model. This model achieved superior performance in glioma grade classification (Accuracy = 0.96, Precision = 1.0, Recall = 0.94, F1-Score = 0.96, AUC = 1.0). Radiomic features derived from peritumoral edema in fused MRI images using the LRD method demonstrated distinct, grade-specific patterns and can be utilized as a non-invasive, accurate, and rapid glioma grade classification method. Full article

Differential privacy (DP) has become a cornerstone for privacy-preserving machine learning, yet its application to high-resolution satellite imagery remains underexplored. Existing DP algorithms, such as DP-SGD, often rely on static noise levels and global clipping thresholds, which lead to slow convergence and poor utility in deep neural networks. In this paper, we propose ADP-SIR, an Adaptive Differential Privacy framework for Satellite Image Recognition with provable convergence guarantees. ADP-SIR introduces two novel components: Convergence-Guided Noise Scaling (CGNS), which dynamically adjusts the noise multiplier based on training stability, and Layerwise Sensitivity Profiling (LSP), which enables fine-grained clipping at the layer level. We provide theoretical analysis showing that ADP-SIR achieves good convergence in non-convex settings under Rényi differential privacy. Empirically, we evaluate ADP-SIR on EuroSAT and RESISC45, demonstrating significant improvements over DP-SGD and AdaClip-DP in terms of accuracy, convergence speed, and per-class fairness. Our framework bridges the gap between practical performance and rigorous privacy for remote sensing applications. Full article

The seed-germination percentage is an important indicator of the seed viability and growth potential and has important implications for plant breeding and agricultural production. Thus, to increase the speed and accuracy in measuring the completion of germination in experimental seed batches for precise germination percentage calculation, we evaluated a You-Only-Look-Once (YOLO)–Seed Germination Detection (SGD) algorithm that integrates deep-learning technology and texture feature-extraction mechanisms specific to germinating seeds. The algorithm was built upon YOLOv7-l, and its applicability was optimised based on the results of our germination experiments. In the backbone network, an internal convolution structure was substituted to enhance the spatial specificity of the initial features. Following the output of the main feature-extraction network, an Explicit Visual Centre (EVC) module was introduced to mitigate the interference caused by intertwined primary roots from germinated seeds, which can affect recognition accuracy. Furthermore, a Spatial Context Pyramid (SCP) module was embedded after enhancing the feature-extraction network to improve the model’s accuracy in identifying seeds of different scales, particularly in recognising small target seeds. Our results with cabbage seeds showed that the YOLO–SGD model, with a model size of 45.22 M, achieved an average detection accuracy of 99.6% for large-scale seeds and 96.4% for small-scale seeds. The model also achieved a mean average precision and F1 score of 98.0% and 93.3%, respectively. Compared with manual germination-rate detection, the model maintained an average absolute error of prediction within 1.0%, demonstrating sufficient precision to replace manual methods in laboratory environments and efficiently detect germinated seeds for precise germination percentage assessment. Full article

Rapid and non-destructive acquisition of soil nutrient information is crucial for precision fertilization and soil quality monitoring. This study aims to establish a Raman spectroscopy-based framework for predicting key soil fertility indicators, including alkali-hydrolyzable nitrogen (AN), total nitrogen (TN), total phosphorus (TP), and organic matter (OM). The framework systematically integrates three typical spectral preprocessing methods (Standard Normal Variate transformation (SNV), Savitzky–Golay first derivative (SG_D1), and wavelet transform (Wavelet)), three feature selection strategies (Recursive Feature Elimination, XGBoost importance, and Random Forest importance), and 14 mainstream regression models to construct a multi-combination modeling system. Model performance was evaluated using five-fold cross-validation, with 80% of samples used for training and 20% for validation in each fold. Preprocessed Raman spectral features served as input variables, while the corresponding nutrient contents were used as outputs. Results showed significant differences in prediction performance across various combinations of preprocessing methods and regression algorithms for the four soil nutrient indicators. For AN prediction, the combination of Raw_SNV preprocessing with ElasticNet and BayesianRidge models achieved the best performance, with Test R² values of 0.713 and 0.721, and corresponding Test NRMSE as low as 0.092. For OM prediction, the same Raw_SNV preprocessing with ElasticNet and BayesianRidge also performed well, yielding Test R² values of 0.825 and 0.832, and Test NRMSE of 0.100 and 0.098, respectively. In TN prediction, both ElasticNet and BayesianRidge under Raw_SNV preprocessing achieved consistent Test R² of 0.74 and Test NRMSE around 0.20, indicating stable reliability. For TP prediction, the BayesianRidge model with Raw_SNV preprocessing outperformed all others with a Test R² of 0.71 and Test NRMSE of just 0.089, followed closely by ElasticNet (Test R² = 0.70, Test NRMSE = 0.092). Overall, the Raw_SNV preprocessing method demonstrated superior performance compared to SG_D1_SNV and Wavelet_SNV. Both BayesianRidge and ElasticNet consistently achieved high R² and low NRMSE across multiple targets, showcasing strong generalization and robustness, making them optimal model choices for Raman spectroscopy-based soil nutrient prediction. This study demonstrates that Raman spectroscopy, when combined with appropriate preprocessing and modeling techniques, can effectively predict soil organic matter and nitrogen in specific soil types under laboratory conditions. These results provide initial methodological insights for future development of intelligent soil nutrient diagnostics. Full article

The compressive strength (Fc) of cement mortar (CM) is a key parameter in ensuring the mechanical reliability and durability of cement-based materials. Traditional testing methods are labor-intensive, time-consuming, and often lack predictive flexibility. With the increasing adoption of machine learning (ML) in civil engineering, data-driven approaches offer a rapid, cost-effective alternative for forecasting material properties. This study investigates a wide range of supervised linear and nonlinear ML regression models to predict the Fc of CM. The evaluated models include linear regression, ridge regression, lasso regression, decision trees, random forests, gradient boosting, k-nearest neighbors (KNN), and twelve neural network (NN) architectures, developed by combining different optimizers (L-BFGS, Adam, and SGD) with activation functions (tanh, relu, logistic, and identity). Model performance was assessed using the root mean squared error (RMSE), coefficient of determination (R²), and mean absolute error (MAE). Among all models, NN_tanh_lbfgs achieved the best results, with an almost perfect fit in training (R² = 0.9999, RMSE = 0.0083, MAE = 0.0063) and excellent generalization in testing (R² = 0.9946, RMSE = 1.5032, MAE = 1.2545). NN_logistic_lbfgs, gradient boosting, and NN_relu_lbfgs also exhibited high predictive accuracy and robustness. The SHAP analysis revealed that curing age and nano silica/cement ratio (NS/C) positively influence Fc, while porosity has the strongest negative impact. The main novelty of this study lies in the systematic tuning of neural networks via distinct optimizer–activation combinations, and the integration of SHAP for interpretability—bridging the gap between predictive performance and explainability in cementitious materials research. These results confirm the NN_tanh_lbfgs as a highly reliable model for estimating Fc in CM, offering a robust, interpretable, and scalable solution for data-driven strength prediction. Full article

Background/Objectives: Acute Lymphoblastic Leukemia (ALL) poses significant diagnostic challenges due to its ambiguous symptoms and the limitations of conventional methods like bone marrow biopsies and flow cytometry, which are invasive, costly, and time-intensive. Methods: This study introduces Neuro-Bridge-X, a novel neuro-symbolic hybrid model designed for automated, explainable ALL diagnosis using peripheral blood smear (PBS) images. Leveraging two comprehensive datasets, ALL Image (3256 images from 89 patients) and C-NMC (15,135 images from 118 patients), the model integrates deep morphological feature extraction, vision transformer-based contextual encoding, fuzzy logic-inspired reasoning, and adaptive explainability. To address class imbalance, advanced data augmentation techniques were applied, ensuring equitable representation across benign and leukemic classes. The proposed framework was evaluated through 5-fold cross-validation and fixed train-test splits, employing Nadam, SGD, and Fractional RAdam optimizers. Results: Results demonstrate exceptional performance, with SGD achieving near-perfect accuracy (1.0000 on ALL, 0.9715 on C-NMC) and robust generalization, while Fractional RAdam closely followed (0.9975 on ALL, 0.9656 on C-NMC). Nadam, however, exhibited inconsistent convergence, particularly on C-NMC (0.5002 accuracy). A Meta-XAI controller enhances interpretability by dynamically selecting optimal explanation strategies (Grad-CAM, SHAP, Integrated Gradients, LIME), ensuring clinically relevant insights into model decisions. Conclusions: Visualizations confirm that SGD and RAdam models focus on morphologically critical features, such as leukocyte nuclei, while Nadam struggles with spurious attributions. Neuro-Bridge-X offers a scalable, interpretable solution for ALL diagnosis, with potential to enhance clinical workflows and diagnostic precision in oncology. Full article

Network trace is a comprehensive record of data packets traversing a computer network, serving as a critical resource for analyzing network behavior. However, in practice, the limited availability of high-quality network traces, coupled with the presence of sensitive information such as IP addresses and MAC addresses, poses significant challenges to advancing network trace analysis. To address these issues, this paper focuses on network trace synthesis in two practical scenarios: (1) data expansion, where users create synthetic traces internally to diversify and enhance existing network trace utility; (2) data release, where synthesized network traces are shared externally. Inspired by the powerful generative capabilities of latent diffusion models (LDMs), this paper introduces NetSynDM, which leverages LDM to address the challenges of network trace synthesis in data expansion scenarios. To address the challenges in the data release scenario, we integrate differential privacy (DP) mechanisms into NetSynDM, introducing DPNetSynDM, which leverages DP Stochastic Gradient Descent (DP-SGD) to update NetSynDM, incorporating privacy-preserving noise throughout the training process. Experiments on five widely used network trace datasets show that our methods outperform prior works. NetSynDM achieves an average 166.1% better performance in fidelity compared to baselines. DPNetSynDM strikes an improved balance between privacy and fidelity, surpassing previous state-of-the-art network trace synthesis method fidelity scores of 18.4% on UGR16 while reducing privacy risk scores by approximately 9.79%. Full article