Search Results (6,610)

Electroencephalogram (EEG)-based seizure prediction has recently emerged as a critical technique for clinical diagnosis and intervention. However, conventional multi-channel EEG analysis methods often overlook the brain’s intrinsic spatial topology and typically employ fixed channel ordering, which constrain their ability to capture cross-regional interactions effectively. To address these limitations, this study proposes a novel Multi-Frequency Topological Neural Network (MF-TopoNet) that jointly captures topological and spatial–temporal characteristics of EEG signals. The proposed framework leverages both constructed functional brain networks and raw multi-channel EEG recordings as inputs, thereby facilitating complementary feature extraction. Specifically, the TopoConv module integrates topological information into the convolutional process and adopts randomized channel fusion to enhance feature diversity. In addition, a cross-band attention mechanism is introduced to model interactions across multiple frequency bands, further improving prediction accuracy. Extensive experiments conducted on the CHB-MIT and Siena datasets demonstrate the superiority and robustness of MF-TopoNet. Under 10-fold cross-validation, the proposed model achieved 95.88% accuracy, 95.60% sensitivity, and 96.15% specificity on the CHB-MIT dataset and 94.01% accuracy, 93.92% sensitivity, and 94.11% specificity on the Siena dataset. These results underscore the importance of incorporating brain topology into deep learning frameworks and highlight the effectiveness of multi-frequency feature fusion for improving seizure prediction performance. Full article

Real estate valuation increasingly requires methodologies that combine predictive accuracy, economic interpretability, and model transparency. This study proposes a Generalized Explainable Hybrid Econometric–Machine Learning Framework integrating econometric modelling, machine learning, and Explainable Artificial Intelligence (XAI) within a unified analytical architecture. The framework is organized around a generalized conceptual valuation equation and implemented through a workflow comprising data quality assessment, econometric benchmarking, hyperparameter optimization, nonlinear machine-learning modelling, five-fold cross-validation, and SHAP-based explainability. The methodology is demonstrated using an internally constructed Azerbaijan commercial real estate dataset and independently assessed using the publicly available UCI residential benchmark dataset through an identical Python-based computational pipeline. Under the complete income-capitalization benchmark, the Artificial Neural Network achieved the highest predictive performance for the Azerbaijan dataset, whereas Random Forest performed best for the external benchmark. A leakage-reduced robustness analysis further showed that the framework retained meaningful predictive capability after excluding Net Operating Income and the Capitalization Rate, with XGBoost providing the strongest performance under the reduced specification. SHAP analysis identified economically meaningful valuation drivers, highlighting income generation in the commercial market and transport accessibility in the residential benchmark. The proposed framework provides a transparent, explainable, and transferable decision-support methodology for modern real estate valuation. Full article

Rapid and accurate diagnosis is crucial for the early detection of malaria caused by Plasmodium parasites. This study primarily focuses on identifying a suitable deep learning model for classifying malaria parasites. In an ablation process, we started with single-branch Convolutional Neural Network (SB-CNN) and dual-branch CNN (DB-CNN) architectures enhanced with attention mechanisms to improve feature representation. Specifically, we incorporate the Convolutional Block Attention Module (CBAM) to focus on both channel-wise and spatially important features. We also use the Efficient Channel Attention (ECA) module to capture local inter-channel relationships, while the Squeeze-and-Excitation (SE) block emphasizes globally significant feature maps, further improving the model’s ability to distinguish between classes. To identify the most effective model, we experimented with EfficientNet-B0 and EfficientNet-B3 as backbone networks and conducted an ablation study by integrating various attention modules, resulting in three DB-CNN variants. We applied t-Distributed Stochastic Neighbor Embedding (t-SNE) to visualize the high-dimensional feature space between infected and uninfected samples. Utilizing the 5-fold cross-validation method on the Thick dataset, the best-performing models included architectures such as SB EfficientNet-B3, SB EfficientNet-B3 combined with CBAM, and DB EfficientNet-B3 across both branches, with CBAM in the first and SE in the second. Full article

Background/Objectives: To achieve effective clinical decision-making from radiographic images, accurate diagnosis of bone fractures is required, but this is difficult due to anatomical variation, subtle fracture appearance, and variable imaging conditions. Existing deep learning studies for fracture detection are predominantly limited to binary classification or single anatomical regions, limiting their real-world clinical utility. This study demonstrates that a single deep learning model can effectively classify multi-region bone fractures by transforming the task into a 22-class classification problem, utilizing the publicly available MultiBoneX dataset. Methods: The proposed model utilizes an EfficientNet-B3 convolutional neural network integrated with a Convolutional Block Attention Module (CBAM) to enhance feature representation by focusing on diagnostically significant areas. We used regularized preprocessing, data augmentation, and structured training to support strong model learning and generalization. Model predictions were interpreted using Gradient-weighted Class Activation Mapping (Grad-CAM) to highlight the image regions that were most important for class selection. Results: When tested on a held-out test set of 3280 images, the model achieved an overall accuracy of 75.03% (95% CI: 73.57–76.43%), with precision, recall, and F1-score of 77.09% (95% CI: 75.28–78.77%), 72.85% (95% CI: 71.00–74.60%), and 73.45% (95% CI: 71.49–75.15%), respectively. The overall multi-class Matthews Correlation Coefficient (MCC) was 0.7361, providing an additional class-imbalance-aware measure of classification performance. Conclusions: These findings demonstrate the feasibility of a unified, multi-class system for diagnosing bone fractures across diverse anatomical sites, providing a scalable foundation for future AI-assisted radiography. Full article

Identifying the small group of customers who hold a given policy in severely imbalanced tabular data is a recurring screening problem in insurance analytics. This study considers binary caravan-insurance policy-ownership classification on the COIL 2000 benchmark, where the positive-class prevalence is below 6%. The benchmark is a single cross-section, so the label describes current ownership rather than a future purchase event. We propose an Attention-based Symmetric AutoEncoder (ASAE) that combines an auxiliary symmetric reconstruction branch, a channel attention gate, and a marginal log-variance regularizer on a 32-dimensional latent representation. The regularizer operates on individual latent variances and is treated as a heuristic rather than as an estimator of joint differential entropy. Under a common tuning and evaluation protocol on a stratified partition, the ASAE is compared with five conventional machine learning methods and seven neural models. Across five paired runs, it achieves an F1-score of 0.6008 ± 0.0115 and an area under the receiver operating characteristic curve (AUC) of 0.8584 ± 0.0034. Relative to TabNet, the strongest baseline considered, the mean differences are 0.064 in F1-score (95% confidence interval 0.043–0.085) and 0.032 in AUC (95% confidence interval 0.017–0.048). The ordering is preserved across five stratified re-splits, four imbalance-handling configurations, and a complete type-consistent preprocessing rerun in which nominal attributes are one-hot encoded, oversampled with SMOTENC, and reconstructed with categorical cross-entropy losses (F1-score 0.6241 ± 0.0074, AUC 0.8702 ± 0.0050). All reported results use stratified random partitions of COIL 2000. Because 27% of the records share an identical predictor vector with another record, the official challenge separation and two grouped partitions are also defined, so that exact-duplicate and sociodemographic overlap can be isolated from the primary split. The training code, split indices, and per-run predictions used for the reported tables are publicly available. Full article

Accurate discrimination among static eccentricity fault (SEFs), dynamic eccentricity fault (DEFs), and mixed eccentricity fault (MEFs) in permanent-magnet-synchronous motors remains difficult because these conditions produce similar stator-current patterns. This paper proposes an STFT–CNN–LSTM-based Eccentricity Fault Diagnosis System (STFT-CL-EFDS) for classifying Normal, SEF, DEF, and MEF conditions using only three-phase stator currents. Six signal transformations were initially compared, after which the three leading representations—STFT, DWT, and CWT—were evaluated with seven neural-network architectures. Sensitivity and ablation analyses selected a 500-sample observation window and a compact log-magnitude STFT representation over the nominal 0–1 kHz band. Following full-schedule retraining, the proposed model achieved 99.57% accuracy and a 99.57% weighted F1-score, with recalls of 100.00%, 98.66%, 100.00%, and 99.53% for Normal, SEF, DEF, and MEF, respectively. It exceeded the strongest machine-learning benchmark, STFT–MLP, by 8.42 percentage points in accuracy and 8.52 percentage points in weighted F1-score. On an independent unseen test bench, the proposed model provided the most balanced response across the three fault types and ultimately converged to the correct class in every case, although temporary DEF–MEF confusion remained. These results demonstrate the effectiveness of STFT-CL-EFDS for current-only multiclass PMSM eccentricity diagnosis. The main contributions of this study are as follows: (1) a systematic comparison of six signal-transformation methods, namely Fast Fourier Transform (FFT), STFT, Discrete Wavelet Transform (DWT), Continuous Wavelet Transform (CWT), Hilbert–Huang Transform (HHT), and Variational Mode Decomposition (VMD), to determine their suitability for eccentricity fault diagnosis; (2) a comparative evaluation of seven neural-network architectures, including CNN, LSTM, CNN–LSTM, DNN, TCN, ModernTCN, and TimesNet, using the three best-performing transformation methods, namely STFT, CWT, and DWT; and (3) the development of a unified current-only STFT-CL-EFDS that combines STFT-based time–frequency representation with convolutional feature extraction and temporal-sequence learning for the classification of Normal, SEF, DEF, and MEF conditions. Full article

This is a narrative review that describes the evolution of terminology and classification schemes of neuroendocrine tumors (NETs), neuroendocrine carcinomas (NECs), mixed neuroendocrine–non-neuroendocrine neoplasms (MiNENs), and paragangliomas of the larynx, with the contribution of immunohistochemistry and treatment of neuroendocrine neoplasms of the larynx. Neuroendocrine neoplasms of the larynx comprise both epithelial (NET and NEC) and neural crest-derived (paraganglioma) neoplasms. Their terminology has evolved substantially over time, with current classifications emphasizing biologically and clinically meaningful categories aligned with contemporary WHO frameworks of other organs. These neoplasms may show overlapping clinical presentation and histomorphological features, which can complicate accurate subclassification. However, precise classification is essential, as these entities display markedly different biological behavior, ranging from indolent to highly aggressive with poor prognosis, and their treatment is essentially histology-tailored. Neuroendocrine neoplasms are classified into three categories comprising six tumor subtypes: well-differentiated neuroendocrine tumors (NETs; grades 1, 2, and 3), poorly differentiated neuroendocrine carcinomas (NECs; small-cell and large-cell types), and paragangliomas. An additional, not yet WHO-recognized category is MiNENs (mixed neuroendocrine–non-neuroendocrine neoplasms), defined by the coexistence of neuroendocrine and non-neuroendocrine components. These tumors exhibit distinct biological behavior and clinical significance. Full article

Normalized Difference Vegetation Index (NDVI) is an important remote sensing indicator for characterizing vegetation growth status and ecosystem changes. Improving NDVI prediction accuracy is of great significance for regional ecological monitoring and conservation. However, existing prediction methods often rely on single meteorological drivers or numerous explanatory variables that are difficult to obtain for future periods, limiting their applicability to pixel-level NDVI forecasting over spatially distributed areas. In this study, a Dual-Factor-Driven Temporal Network (DfT-Net) was proposed for pixel-level NDVI prediction in the Hulunbuir grassland based on MODIS NDVI data and ERA5-Land temperature and precipitation data from 2013 to 2024. The model was independently applied to valid 1000 m grassland pixels, with the same model parameters shared across pixels. This pixel-wise prediction strategy enables the model to learn common meteorological–vegetation response patterns across different pixels while generating spatially distributed NDVI predictions. For temporal feature modeling, Time Series Decomposition (TSD) was first applied to the temperature and precipitation sequences to decompose them into trend, seasonal, and residual components, thereby characterizing their multi-scale temporal variations and recurrent seasonal patterns. Subsequently, one-dimensional convolutional neural networks (CNNs) and long short-term memory networks (LSTMs) were employed to extract local temporal patterns and long-term temporal dependencies, respectively. Experimental results showed that DfT-Net achieved an RMSE of 0.100, an MAE of 0.074, and an R² of 0.828 on the independent temporal test set (2022–2024). Under the same experimental setting, DfT-Net achieved the lowest RMSE and MAE and the highest R² among the evaluated models, including LSTM, CNN, TSD-CNN, TSD-LSTM, and CNN-LSTM. These results indicate that DfT-Net effectively integrates dual-factor meteorological driving, multi-scale temporal feature representation, and pixel-level prediction, providing a useful framework for spatially distributed grassland NDVI forecasting and ecological monitoring. Full article

Bone marrow lesion (BML) volume is an essential biomarker for understanding knee osteoarthritis (KOA). However, automatic BML segmentation remains challenging due to the irregular shapes and indistinct boundaries of these lesions in knee magnetic resonance images (MRI). To improve BML segmentation, this study investigated two established strategies in the specific context of BML segmentation: (1) integrating bone segmentation as an additional output channel in deep neural networks to facilitate BML segmentation, and (2) incorporating test-time augmentation (TTA) to reduce uncertainty during testing. The added bone segmentation channel provides auxiliary anatomical information that may facilitate BML localization. TTA was used to improve boundary alignment and reduce false positives by generating more robust predictions. Multiple State-of-the-Art deep neural networks for segmentation were employed as the baseline models to compare performance before and after implementing the proposed strategies. A 10-fold cross-validation was conducted on a dataset of knee MR scans from 300 participants. Segmentation performance was evaluated using the Dice similarity coefficient (DSC) for overlap accuracy and the 95% Hausdorff Distance (HD95) for boundary alignment. Paired t-tests were used to assess the significance of improvements from the proposed strategies. Both strategies produced improvements in segmentation performance, although the magnitude and statistical significance of the improvements varied across architectures. The DSC improved from 63.1% to 64.8% for Residual U-Net, 64.2% to 65.8% for Swin UNETR, 61.5% to 66.5% for Attention U-Net, and 66.6% to 69.0% for UNet++. These gains were accompanied by improvements in boundary accuracy and reductions in false positives, reflected in lower HD95 values. Comparison with additional medical image segmentation models under the same evaluation framework showed that the dual-channel UNet++ with TTA achieved the highest BML DSC of 69.0%, followed by U-Mamba at 68.6% and nnU-Net at 65.2%, while U-Net + InceptionResNet-v2 achieved 57.7%. These findings support the potential value of dual-channel and TTA strategies for automated BML analysis, while further validation on independent datasets is needed to assess their broader generalizability and clinical utility. Full article

Brown Rot (Monilinia spp.) and Leaf Curl (Taphrina deformans) are principal fungal diseases affecting peach (Prunus persica L. Batsch) production, lacking validated AI-based diagnostic tools in tropical highland orchards. This study presents a compact Convolutional Neural Network (3658 trainable parameters), applied identically to fruit and leaf classification, integrated with a background-removal preprocessing pipeline and evaluated through stratified 5-fold cross-validation on 800 in-situ images from four orchards in Cómbita and Choachí, Colombia. The proposed architecture achieved mean accuracies of 82.8% (fruit) and 95.3% (leaf), with AUC values of 0.87 and 0.98, and a trained model footprint of approximately 100 KB, supporting storage- and bandwidth-efficient deployment. Benchmarked against ImageNet-pretrained MobileNetV3-Small and MobileNetV2 under an identical protocol, the proposed architecture matched or exceeded MobileNetV3-Small on leaf classification despite a 257-fold smaller parameter count, and achieved comparable or lower inference latency than both larger backbones. To our knowledge, this is the first validated system for simultaneous detection of both pathogens in Prunus persica under real field conditions, combining a compact, deployment-ready architecture with an ablation-verified preprocessing pipeline. The proposed model was deployed in the DurAPP web platform, giving peach growers in tropical highland regions a practical, low-footprint diagnostic tool suited to smallholder farming conditions. Full article

In recent years, U-Net-based architectures have been widely applied to polarimetric synthetic aperture radar (PolSAR) semantic segmentation. However, successive downsampling may lead to the loss of fine spatial details, while conventional U-Net-style decoder directly concatenates encoder features with the corresponding decoder features without explicitly accounting for their semantic discrepancy, potentially introducing redundant or irrelevant information and weakening feature discrimination. To address these limitations, this paper proposes a lightweight complex-valued high-resolution U-Net (CV-HRU-Net) with a complex-valued cross-gated attention (CV-CGA) module for PolSAR semantic segmentation. CV-HRU-Net employs a complex-valued high-resolution network (CV-HRNet) as the encoder to maintain high-resolution representations through parallel multi-resolution streams, while a complex-valued U-Net (CV-U-Net) decoder progressively incorporates multi-resolution high-level semantic features for pixel-wise prediction. To improve encoder–decoder feature interaction, CV-CGA adaptively calibrates the core decoder features using encoder information. Specifically, CV-CGA integrates the Convolutional Block Attention Module, Transformer-style cross-attention with decoder features as queries and encoder features as keys and values, and adaptive gated recalibration to enhance semantic selectivity and boundary representation. Experiments on two airborne and two spaceborne PolSAR datasets demonstrate that the proposed network achieves accurate land-cover segmentation and precise boundary delineation by jointly exploiting polarimetric phase relationships, fine spatial details, and multi-resolution semantic information. Furthermore, CV-CGA substantially improves segmentation accuracy and boundary F1 scores while introducing only marginal model-size overhead. Full article

This paper proposes a convolutional neural network (CNN)-based image recognition method for crack detection in masonry walls to improve accuracy and efficiency. A sample database is constructed using laboratory-captured images of cracked masonry walls, augmented through rotation, mirroring, gamma correction, contrast adjustment, and brightness modification. A dual-scale CNN architecture is designed by integrating the characteristics of VGG-16 and GoogleNet, enabling multi-scale feature extraction for load-bearing masonry wall images. The network adopts two input scales: a modified VGG-16 branch with 100 × 100 inputs and an optimized GoogleNet branch with 50 × 50 inputs. Both branches undergo feature extraction and classifier training. Leveraging transfer learning, the model is pre-trained on the large-scale ImageNet dataset and fine-tuned using our self-built masonry wall database. Multi-scale features from dual branches are weighted and fused to generate final crack recognition results. Experimental results demonstrate a training accuracy of 94.33%, precision of 94.78%, and F1-score of 94.68%, outperforming traditional CNN models. The proposed method achieves high recognition accuracy for masonry load-bearing wall cracks and provides reliable technical support for practical engineering applications. Full article

Wetlands monitoring is essential for sustainable watershed biodiversity conservation, and climate resilience. This study performs a spatiotemporal wetland change analysis for the Sangamon River Watershed, Illinois between 2000 and 2025, using Landsat 5 Thematic Mapper (TM), Sentinel–2 Surface Reflectance (SR), Synthetic Aperture Radar (SAR), Gray–Level Co–occurrence Matrix (GLCM) and terrain data through cloud–based processing in Google Earth Engine (GEE), Google Colab and ArcGIS Pro 3.6.0. We conducted a comparative assessment of deep learning (Deep Neural Network [DNN], U-Net, Attention U-Net, and SegFormer), and machine learning models (Random Forest [RF], Gradient Tree Boosting [GTB], and Support Vector Machines [SVM]) through pixel–based and object–based methods. National Land Cover Database (NLCD) was used for training and validation using stratified random sampling for five categories namely wetlands, forest, agriculture/grassland/barren land, urban/developed and water. A proportion of 54.85% (860.68 km²) of wetlands extent was lost to other land uses, particularly agriculture, urban and forest, along with 46.51% (694.88 km²) forest and 36.80% (66.13 km²) water bodies loss. Agriculture/grassland/barren and urban/developed witnessed increases of 8.56% (820.62 km²) and 72.92% (799.52 km²), respectively. For Landsat–based and Sentinel–based classifications, SegFormer outperformed all ML and DL classifiers with (OA = 94%, Kappa = 0.89, mean F1 = 0.80, mean IoU = 0.70 and OA = 96%, Kappa = 0.92, mean F1 = 0.95, mean IoU = 0.73, respectively) with excellent wetland delineation (PA = 0.99, UA = 0.97, F1 = 0.98, IoU = 0.97 and PA = 0.99, UA = 0.99, F1 = 0.98, IoU = 0.99, respectively). Sentinel–based classifications had improved performance than Landsat, while object–based models consistently outperformed pixel–based methods. The Digital Elevation Model (DEM) and slope were the most influential predictors for RF models, while GLCM and SAR produced negligible influence. The integrated and comparative GeoAI framework provides a robust methodology for watershed–scale wetland monitoring and supports evidence–based conservation, restoration prioritization, climate resilience, and sustainable land–use planning, while offering strong potential for application in other agricultural watersheds following regional validation. Full article

The standard 12-lead ECG is the gold-standard tool for cardiac diagnosis and monitoring; however, its multi-electrode configuration limits its use in prolonged monitoring and wearable devices. Existing portable ECG devices typically record only a single lead, reducing their diagnostic value. To overcome this limitation, this study proposes novel Multi-Input Multi-Output neural networks (MIMO-Nets) that reconstruct a standard 12-lead ECG from a single raw ECG lead without requiring any feature extraction techniques. Four deep learning architectures were investigated: MIMO-Conv-Net, MIMO-U-Net, Conv-Net, and U-Net. Each model was trained and evaluated using all 12 standard ECG leads with different length L as candidate inputs to determine the optimal single-lead acquisition strategy. Results show that Lead II with L = 256 consistently provides the highest reconstruction performance across all architectures, making it the most informative single-lead input. Using Lead II as input, MIMO-U-Net yielded the highest reconstruction accuracy on a processed clinical dataset (ST-Petersburg INCART), with an average correlation of 0.864 at L = 256, closely followed by U-Net (0.862). On raw, unprocessed recordings from the PTB-XL database, both models achieved lower but still competitive correlations (0.740 and 0.732, respectively). These findings demonstrate the feasibility of single-lead-to-12-lead ECG reconstruction and highlight the role of signal preprocessing in reconstruction fidelity, supporting the development of wearable ECG systems for continuous cardiac monitoring. Full article

Granulation is a major physiological disorder that compromises sweet orange quality. Conventional destructive detection methods lead to food waste and cannot be used for batch inspection. This study proposes a nondestructive approach for detecting granulation in sweet oranges using air-jet transient excitation and a laser Doppler vibrometer (LDV). Acoustic vibration spectra were acquired from 640 sweet orange samples. Using both competitive adaptive reweighted sampling (CARS)-extracted feature parameters and raw acoustic vibration spectra as inputs, an ISNet-1D model integrating a multi-scale Inception module and a squeeze-and-excitation (SE) attention mechanism was developed, and its performance was compared against those of partial least squares discriminant analysis (PLS-DA), support vector machine (SVM), K-nearest neighbor (KNN), random forest (RF), and baseline deep learning models including one-dimensional convolutional neural network (1D-CNN), Visual Geometry Group network 16 (VGG16), and residual network v1 (ResNet-v1). The results demonstrated that the ISNet-1D model trained on the full raw vibration spectrum achieved the best performance, with an overall test set accuracy, recall, and specificity of 92.97%, 95.00%, and 91.18%, respectively. Ablation experiments revealed that removal of the Inception branches and the SE module reduced the overall test accuracy by 5.47% and 4.69%, respectively, indicating that their combination effectively extracts multi-scale acoustic vibration features and enhances model precision. Gradient-weighted class activation mapping further identified the critical frequency bands primarily relied upon by the model for prediction. Collectively, noncontact acoustic vibration detection combined with ISNet-1D provides a viable method for nondestructive granulation detection in sweet oranges. Full article