Search Results (8,743)

Mixed-signal simulation is indispensable for verifying modern integrated circuits that tightly couple analog and digital subsystems, yet the field lacks a unified framework for systematically comparing its diverse methodologies. This paper addresses that gap by proposing a novel three-axis taxonomy that classifies simulation methods along abstraction level, solver methodology, and analysis type, together with a comparative evaluation framework based on five quantitative metrics: accuracy, throughput, capacity, convergence reliability, and scalability. Applying this framework, we systematically compare thirteen classical method categories—spanning SPICE, FastSPICE, RF/periodic steady-state, behavioral modeling, co-simulation, and model order reduction—and eight AI/ML approaches including Gaussian process surrogates, graph neural networks, physics-informed neural networks, Bayesian optimization, and reinforcement learning. Our analysis reveals a clear maturity stratification: classical methods remain the only signoff-accurate approaches, Bayesian optimization represents the most industrially validated AI contribution with integration across all three major EDA platforms, while Neural ODE solvers and LLM-based design tools remain at the research stage. We identify a persistent academic-to-industry gap driven by foundry model complexity, limited benchmark diversity, and topology-specific overfitting. The proposed taxonomy and comparative framework provide practitioners with structured guidance for simulation method selection and highlight specific research directions needed to bridge the gap between AI promise and industrial deployment. Full article

Internationally, there are increased pressures for primary schools to meet academic curriculum outcomes primarily driven by performance metrics and targets. Sitting alongside this context are competing concerns for the decline in children’s play opportunities to bolster their overall health and wellbeing. Adopting play-based pedagogies in primary schools can infuse more play into children’s lives whilst meeting curriculum outcomes. Despite the perceived importance of play during childhood, play-based pedagogies are still mostly positioned as legitimate pedagogical approaches in prior to school settings. Given this landscape, this research seeks to understand contemporary educational research of play-based pedagogies in primary schools by conducting a scoping review. Through presenting a narrative account of the literature, and synthesising these ideas into broader themes, the research identified that there remains international interest in play-based pedagogies in the primary years of school but despite this, questions surrounding its legitimacy remain. This review and subsequent discussion surface potential next steps including a recommendation to increase empirical research on the adoption of play-based pedagogies in schools with consideration of using a ’Mosaic approach’ to data collection, as well as research focusing on the active and intentional role of the teacher. Lastly, as a way forward, the research brings to light the potential of creating a ‘space’ for the merging of two knowledge systems from two often siloed approaches to education—early childhood and primary—to create a new pathway. Such a pathway has potential to support continuity of learning, student engagement, children’s health, and wellbeing. Full article

With increasing port automation and operational intensity, the gearboxes of gantry cranes widely used in bulk cargo terminals are prone to bearing and gear failures under prolonged heavy loads, intense vibrations, and complex operating conditions. Since fault samples often exhibit imbalanced distributions, this imposes two higher requirements on diagnostic methods—first, the ability to effectively address sample imbalance and, second, the capability to simultaneously identify multiple fault categories. To address these challenges, this paper proposes a joint diagnostic method integrating an improved Conditional Wasserstein Generative Adversarial Network with Gradient Penalty (CWGAN-GP) and Multi-Task Learning (MTL). First, the modified CWGAN-GP performs conditional augmentation for minority fault classes, evaluating synthetic sample authenticity and diversity through multiple metrics. Subsequently, a multi-channel diagnostic network is constructed, in which vibration signals are fed into two parallel sub-networks: time–frequency features are extracted from the Short-Time Fourier Transform (STFT)-based time–frequency representations via a residual-block Convolutional Neural Network (CNN), while temporal features are captured from the raw time-domain signal using a Bidirectional Long Short-Term Memory (Bi-LSTM) with an attention mechanism. An attention fusion layer then integrates these two feature types, enabling joint classification of bearings and gears within a multi-task learning framework. Experimental validation on public gearbox datasets and port gantry crane gearbox datasets demonstrates that this method achieves an average diagnostic accuracy exceeding 97%. The proposed method reduces the impact of class imbalance, thereby improving the accuracy and stability of multi-task fault identification. Full article

This study addresses the automatic classification of cells in electroluminescent panel images to detect photovoltaic cell defects. The images used in the study were obtained from a solar panel production line. An original dataset consisting of 37,538 cell images with eight defect classes (Cell-Interconnection, Electrically Insulated Cell Parts, Finger Defect, Material, Microcrack, Multi-Defect, Normal, Visual) was prepared by applying RLSA-based automated cell segmentation enhanced with morphological processing to the photovoltaic panel images. A novel CNN–Transformer model with a self-attention mechanism, called PVELNet, is proposed for classifying defect types. Experimental studies were conducted with 16 deep learning models to compare the proposed model. F1-Score, Precision, Recall, and Accuracy evaluation metrics were used in the experimental study. Furthermore, the Confusion Matrix results obtained from the 16 deep learning models and the proposed PVELNet model are presented. The results were obtained using a relatively balanced dataset prepared for this study. PVELNet achieved 95.71% accuracy, outperforming other models. With 1.79 million parameters and a memory requirement of 46.1 MB, the PVELNet model is relatively lightweight. As a result, it demonstrates the potential to control processes on actual solar panel production lines. Full article

Synthetic Aperture Radar (SAR) and optical image fusion technology plays a crucial role in remote sensing applications. It effectively combines the high spatial resolution and rich spectral information of optical images with the all-weather and penetrating observation advantages of SAR images, thereby significantly enhancing image interpretation accuracy and task execution capabilities. This paper systematically reviews deep learning-based fusion methods for SAR and optical images, with a particular focus on recent advances in deep learning models. Furthermore, it summarizes commonly used evaluation metrics for assessing fusion image quality, providing a basis for comparing and analyzing the performance of different methods. In addition, commonly used SAR-optical fusion datasets are briefly reviewed to highlight their roles in algorithm development and performance evaluation. Unlike conventional review articles, this paper further analyzes the guidance and supporting role of fusion algorithms from the perspective of typical and specific applications. Finally, it identifies key challenges and issues faced by current fusion methods, including data registration, model lightweight design, and multimodal feature alignment, and offers perspectives on future research directions. This review aims to provide routes and references for the development of SAR and optical image fusion technology. Full article

Background/Objectives: Stroke generates 157 million disability-adjusted life-years (DALYs) annually, making it the leading neurological cause of global disease burden. Artificial intelligence (AI) and machine learning (ML) have emerged as transformative technologies across the stroke care continuum. This narrative review maps the trajectory of AI in stroke medicine over the decade from 2015 to 2025. Methods: We conducted a narrative review with a structured, pre-specified search strategy across eight pre-specified thematic clusters using PubMed/MEDLINE (January 2015–December 2025), identifying 8549 records and including 1335 studies after screening. Inclusion criteria encompassed primary research articles, systematic reviews, meta-analyses, and RCTs reporting quantitative performance metrics or clinical outcome data for AI/ML in stroke. Results: Stroke imaging AI is the most commercially mature domain, with over 30 FDA-cleared tools. Automated ASPECTS scoring reduced radiologist reading time by 74.8% (AUC 84.97%; 95% CI: 83.1–86.8%). The only triage AI RCT demonstrated an 11.2 min reduction in door-to-groin time without significant improvement in 90-day functional independence (OR 1.3, 95% CI 0.42–4.0). Brain–computer interface rehabilitation showed significant upper limb recovery in a 17-center RCT (FMA-UE mean difference +3.35 points, 95% CI 1.05–5.65; p = 0.0045). AF detection AI is FDA-cleared and RCT-validated. LLMs and federated learning are pre-regulatory but growing exponentially. Conclusions: AI in stroke has achieved diagnostic maturity but therapeutic immaturity. Bridging algorithmic performance to patient outcomes, addressing equity gaps, and building the economic evidence base for scalable deployment are the defining challenges of the next decade. Full article

Neurotoxicity from drugs and environmental pollutants poses serious risks to brain function, yet existing computational models mainly target general neurotoxicity and lack specialized tools for brain-specific assessment. This study aimed to develop and validate a high-performance, brain-focused neurotoxicity prediction framework to improve drug safety evaluation and toxicity screening. We systematically analyzed molecular features, clustering patterns, and target predictions of brain-toxic compounds. Multiple feature representations were compared, including traditional molecular fingerprints, knowledge-guided pre-trained graph Transformer (KPGT) embeddings, and transformer-based MolFormer embeddings, combined with machine learning classifiers. Model performance was evaluated using multiple metrics, and SHAP analysis was conducted to identify influential molecular substructures. Toxic molecules showed physicochemical properties favoring central nervous system (CNS) penetration, including lower molecular weight, lower LogP, fewer hydrogen bond donors/acceptors, fewer rotatable bonds, and lower polar surface area (PSA). The KPGT-MLP model achieved the best balanced performance, with an accuracy (ACC) of 0.8928 and an ROC-AUC of 0.9459, clearly outperforming traditional fingerprint-based models, MolFormer-based models, and general prediction tools such as DI-NeuroT and ADMETlab 3.0. Overall, this study establishes a robust framework for brain-specific neurotoxicity prediction, with the KPGT-MLP model demonstrating strong accuracy and robustness. The proposed approach provides an effective strategy for early neurotoxicity screening and risk assessment, offering valuable insights for safer drug design and advancing computational toxicology and drug discovery. Full article

High-voltage bushings, the fault-prone key electrical components of transformers, are critical for real-time and high-accuracy fault monitoring and management. Intelligent fault detection via infrared images is plagued by low classification accuracy due to massive interference from similar tubular objects and small target characteristics. This study proposes a lightweight deep learning model, MTrip–YOLO, an improved YOLO11n integrated with Mamba-2 (Structured State Space Duality, SSD) and Triplet Attention, to achieve efficient fault monitoring in complex backgrounds. The training and validation dataset comprises open-source images, on-site data from a substation, and field-collected infrared images, categorized into four types: normal bushings, poor contact, oil shortage, and high dielectric loss faults. Mamba-2 captures the long-range global context of infrared features with its linear-complexity long-range modeling capability to enhance feature extraction, while Triplet Attention suppresses complex background radiation noise through cross-dimensional interaction without dimensionality reduction, enabling the model to focus on small targets and accurately classify bushings from morphologically similar strip-shaped objects. Experimental results show that MTrip–YOLO achieves a top mAP50 of 91.6% and a minimal parameter count of 1.9 M, outperforming Faster R-CNN, RT-DETR, and YOLO26n across all evaluated metrics and being potentially suitable for edge deployment on UAV-mounted or handheld infrared platforms, pending hardware validation on embedded computing devices. Ablation experiments verify the independent contributions of Mamba-2 (0.8027% mAP50 improvement) and Triplet Attention (0.89327% mAP50 improvement), with a synergistic effect from their combination. MTrip–YOLO provides a potential edge-deployable solution for high-voltage bushing fault monitoring, offering important application value for the intelligent operation and maintenance of substations. Full article

Hydroxylamine-modified transition-metal oxides (HA-TMOs) represent a promising class of catalysts for activating peroxymonosulfate (PMS) to degrade antibiotics. However, identifying energy-efficient operational conditions remains challenging. This study established a comprehensive dataset encompassing 600 experimental records from both in-house experiments and literature and systematically compared 12 machine learning algorithms for predicting the antibiotic degradation efficiency (%) in hydroxylamine-modified transition metal oxide/peroxymonosulfate (HA-TMO/PMS) systems. Among these models, CatBoost delivered the best generalization (test-set R² = 0.8110, RMSE = 8.92, MAE = 6.15) across repeated 80/20 stratified splits with 5-fold cross-validation, outperforming other ensembles as confirmed by cumulative distribution plots and error-metric analyses. Moreover, the permutation importance analysis identified PMS dosage, HA level, pH, initial pollutant concentration, and catalyst loading as the dominant drivers governing the pollutant removal performance. The partial-dependence plots, incorporating two-variable interactions, elucidated the response surfaces and enabled the discovery of operating windows that jointly maximize degradation efficiency and minimize electrical energy per order (EE/O). ML-guided optimization yielded optimal conditions, which were experimentally verified with sulfamethoxazole (SMZ). The HA-Co₃O₄/PMS system achieved the highest degradation rate constant (k = 0.11 min⁻¹) and the lowest EE/O value (6.84 kWh·m⁻³·order⁻¹), markedly improving kinetics and reducing energy consumption compared with non-optimized runs. This work establishes an interpretable ML framework that connects catalyst properties and reaction conditions to degradation kinetics and mechanisms, providing a practical strategy for the screening and scale-up of energy-efficient HA-TMOs/PMS-based advanced oxidation processes (AOPs). Full article

This paper introduces the Polynomial Perceptron (PP), a structured extension of the classical perceptron that incorporates explicit polynomial feature expansions to model nonlinear interactions while preserving analytical transparency. By expressing feature interactions in closed functional form, PP captures higher-order dependencies through a compact set of learned coefficients, establishing a principled trade-off between expressivity and parameter efficiency. The proposed architecture is evaluated across heterogeneous domains, including text, image, and structured data tasks, under controlled experimental settings with parameter-matched baselines. Performance is assessed using standard metrics such as classification accuracy and model complexity (parameter count). Empirical results demonstrate that low-degree PP models achieve competitive accuracy compared to multilayer perceptrons and convolutional neural networks, while requiring significantly fewer parameters. An ablation study further analyzes the impact of polynomial degree on predictive performance, revealing diminishing returns beyond moderate degrees and highlighting favorable efficiency–accuracy trade-offs. A key advantage of PP lies in its intrinsic interpretability. Unlike conventional deep learning models that rely on post hhoc explanation methods, PP provides direct analytical insight through its explicit polynomial structure, enabling decomposition of predictions into feature-, token-, or patch-level contributions without surrogate approximations. Overall, the results indicate that PP offers a lightweight, interpretable, and computationally efficient alternative to standard neural architectures, particularly well-suited for resource-constrained environments and applications where transparency is critical. Full article

Rolling bearing fault diagnosis under complex and noisy operating conditions requires not only high diagnostic accuracy but also interpretability that can be quantitatively verified against physically meaningful excitation structures. However, many existing deep learning approaches rely on a single time–frequency (TF) representation and provide limited, non-verifiable links between model decisions and the original vibration patterns. To address this issue, we propose MBT-XAI, a multi-wavelet TF fusion network with a Token-to-Spectrum Traceback (TST) mechanism for structure-preserving, physics-consistent interpretability. Three complementary wavelets, namely Morlet, Mexican Hat, and Complex Morlet, are used to construct multi-view TF representations, which are encoded into RGB channels and adaptively fused via cross-channel attention within a Transformer backbone. TST maps patch-token attributions back to the TF domain, enabling quantitative evaluation of physics consistency through overlap-based metrics. Experiments on the public CWRU dataset and an industrial IMUST dataset show that MBT-XAI achieves 98.13 ± 0.24% and 96.23 ± 0.31% accuracy at SNR = 0 dB, outperforming the strongest baseline by 2.83% and 2.43%, respectively. Under AWGN contamination, MBT-XAI maintains 95.44 ± 0.38%/93.45 ± 0.47% accuracy on CWRU and 95.80 ± 0.33%/92.91 ± 0.51% accuracy on IMUST at SNR = −2/−4 dB. Under colored-noise contamination, the proposed method also preserves robust performance under pink and brown noise at the same SNR levels. Quantitative interpretability evaluation further indicates high alignment between salient frequency regions and theoretical fault-characteristic bands, with IoU = 80.21 ± 0.86% and Coverage = 91.70 ± 0.63%. In addition, MBT-XAI requires 10.393 M parameters and 10.678 GFLOPs, with an inference latency of 14.7 ms per sample (batch size = 1) on an NVIDIA GeForce RTX 3060 GPU. These results suggest that multi-wavelet TF modeling with attention-based fusion and TF-level traceback provides an accurate, robust, and physics-consistent framework for intelligent bearing fault diagnosis. Full article

In highly digitalized and networked modern trading systems, large volumes of heterogeneous data are continuously generated from multiple sources during market operations. However, due to the complexity of data structures, significant differences in temporal scales, and constraints imposed by data privacy protection, traditional single-source modeling approaches are unable to fully exploit multisource information. To address this issue, a federated multimodal prediction framework for complex market systems, termed Federated Market-Sensor Transformer (FMST), is proposed. In this framework, data originating from different information sources are uniformly modeled as multimodal time series. A multimodal market-sensor representation module is constructed to perform unified feature encoding, and a cross-modal Transformer fusion architecture is employed to characterize dynamic interaction relationships among different information sources. Meanwhile, a federated collaborative learning mechanism is introduced during the training phase, enabling multiple data nodes to perform collaborative model optimization without sharing raw data. In this manner, data privacy can be preserved while improving the cross-region generalization capability of the model. Systematic experimental evaluation is conducted on the constructed multimodal market-sensor dataset. The experimental results demonstrate that the proposed method consistently outperforms traditional statistical models and deep learning approaches across multiple evaluation metrics. In the main prediction experiment, FMST achieves a root mean square error (RMSE) of $0.1136$, a mean absolute error (MAE) of $0.0832$, and a coefficient of determination $R^2$ of $0.8517$, while the direction prediction accuracy reaches $74.56\%$, clearly outperforming baseline models including ARIMA, LSTM, Temporal CNN, Transformer, and FedAvg-LSTM. In the cross-region generalization experiment, FMST maintains strong performance, achieving an RMSE of $0.1242$, an MAE of $0.0908$, an $R^2$ value of $0.8261$, and a direction prediction accuracy of $72.48\%$. The ablation study further indicates that the three core components—multimodal market-sensor representation, cross-modal Transformer fusion, and federated collaborative learning—each make important contributions to the overall model performance. These experimental findings demonstrate that the proposed method can effectively integrate multisource market information and significantly enhance the prediction capability for complex market dynamics, providing a new technical pathway for the application of artificial intelligence-driven multimodal sensing systems in economic data analysis. Full article

The hole expansion ratio (HER) is a critical formability metric for advanced high-strength steels (AHSS) in automotive applications; however, its experimental determination is costly and time-consuming. This study presents a machine learning framework for HER prediction using physics-informed synthetic data generation to address data scarcity challenges. A dataset of 300 AHSS conditions was generated based on validated empirical relationships from the literature, incorporating chemical composition, microstructure fractions, and mechanical properties. Multiple machine learning algorithms were evaluated, with the optimized Gradient Boosting model achieving excellent predictive performance on an independent test set (R² = 0.80, RMSE = 5.81%, MAE = 4.93%). The feature importance analysis revealed physically meaningful rankings, with the ultimate tensile strength dominating (40.9%), followed by the bainite volume fraction (15.1%), martensite volume fraction (14.7%), and strain hardening exponent (12.4%). These rankings align with the established metallurgical understanding, thereby validating our synthetic data approach. The results demonstrate that machine learning models trained on physics-informed synthetic data can accurately predict the HER values with errors comparable to the experimental variability, providing a practical tool for accelerated AHSS design and optimization in automotive applications. Full article

Direct urea–hydrogen peroxide fuel cells (DUHPFCs) are promising for sustainable power generation, but their performance is governed by highly nonlinear material and operating interactions. This study develops a machine-learning framework employing a multi-output artificial neural network (ANN) to predict cell voltage, power density (PD), and substrate-based energy efficiency (SEE) of DUHPFCs. The ANN exhibits excellent predictive accuracy, achieving coefficients of determination (R²) above 0.995 and normalized root mean square errors (NRMSE) below 1.75 × 10⁻² for all outputs. Model interpretability is enhanced by using Shapley additive explanations and partial dependence plots, which identify current density as the dominant factor affecting DUHPFC performance, followed by temperature and anolyte composition. The ANN is coupled with a multi-objective Pareto-search algorithm optimization (PAO) to resolve the trade-offs among competing performance metrics. Under different optimization objectives, a DUHPFC with an Ni_0.2Co_0.8/Ni-foam anode is predicted to achieve a maximum PD of 45.6 mW/cm² with a low SEE of 2.6% or a maximum SEE of 15.2% with a moderate PD of 40.9 mW/cm². Additionally, a balanced operating regime is identified, achieving a PD of 43.1 mW/cm² and an SEE of 13.9%. Overall, the proposed framework provides an effective decision-support tool for optimizing DUHPFC performance under competing objectives. Full article

Multivariate time series anomaly detection holds critical application value in key domains such as industrial system monitoring, financial risk management, and medical surveillance. However, existing approaches face two major challenges: reconstruction-based or prediction-based models tend to adapt to anomalous patterns during training, thereby weakening the distinction between normal and abnormal samples; furthermore, the non-stationary nature of time series leads to distribution shifts between training and testing data, impairing model generalization. To address these issues, this paper proposes the TFCID model. The model innovatively leverages diffusion principles to effectively impute missing time series data while capturing significant frequency-domain features. In the temporal processing stream, an unconditional diffusion model combined with imputation masking is employed to achieve high-precision imputation of randomly missing values, effectively preventing anomalies from interfering with model training. In the frequency-domain processing stream, an amplitude-aware frequency-domain masked autoencoder is introduced to specifically capture periodic or trend-based pattern anomalies. The model mitigates distribution shift by constraining the discrepancy between temporal and frequency-domain representations via adversarial contrastive learning, and uses this discrepancy as a robust anomaly scoring metric. Experimental results on five public benchmark datasets show that TFCID significantly outperforms state-of-the-art methods in detection accuracy (F1-Score), validating its effectiveness in anomaly detection tasks. Full article