Search Results (3,443)

Deep learning has achieved expert-level performance in medical imaging analysis. However, models often fail to generalize across patient populations due to cross-population domain shifts, distributional differences arising from demographic variability, variations in imaging protocols, scanner hardware, and differences in disease prevalence. This challenge limits the real-world deployment and can increase health inequities. This review systematically examines the nature, causes, and impact of cross-population domain shift in deep learning-based medical imaging analysis. We analyzed 50 peer-reviewed studies from 2020 to 2025, evaluating the proposed methodologies for handling population shifts, the datasets employed, and the metrics used to assess performance. Our findings demonstrate that performance degradation ranged from 10–25% when models were tested on unseen populations, emphasizing the substantial impact of domain shifts on model generalizability. The literature reveals that mitigation strategies broadly fall into two categories: data-centric approaches, such as augmentation and harmonization, and model-centric approaches, including domain adaptation, transfer learning, adversarial learning, multi-task learning, and continual learning. While domain adaptation and transfer learning are the most widely used, their performance gains across populations remain modest, ranging from 5–15%, and are not supported by external validation. Our synthesis reveals a significant reliance on large, publicly available datasets from limited regions, with an underrepresentation of data from low- and middle-income countries. Evaluation practices are inconsistent, with few studies employing standardized external test sets. This review provides a structured taxonomy of mitigation techniques, a refined analysis of domain shift characteristics, and an in-depth critique of methodological challenges. We highlight the urgent need for more geographically and demographically inclusive datasets, adaptable modeling techniques, and standardized evaluation protocols to enable accurate and equitable AI-driven diagnostics across diverse populations. Finally, we outline future research directions to guide the development of robust, generalizable, and fair models for medical imaging analysis. Full article

Artificial-intelligence-driven wayside monitoring has become a promising solution for early identification of railway wheel flats, enabling safer operations and more efficient maintenance planning. This study introduces a comparative investigation of supervised and unsupervised machine learning strategies for wheel flat identification, with particular emphasis on real-time applicability and sensor cost reduction. Support Vector Machines (SVMs) and k-means clustering are evaluated as representative supervised and unsupervised approaches using vibration data obtained from numerically simulated train–track interactions under realistic operating conditions, including train speeds of 120 km/h and 200 km/h and multiple wheel flat severities. A key contribution of this work is the proposal of a simplified supervised classification framework that directly exploits Auto-Regressive features extracted from rail-mounted accelerometers, eliminating the need for feature normalization and multi-sensor data fusion. This simplification significantly reduces computational effort, making the approach suitable for real-time deployment in operational railway environments. In parallel, a systematic sensitivity analysis is conducted to assess the influence of sensor placement and to identify the minimum sensor configuration required to achieve reliable damage classification. The outputs from the current study show that an SVM emerges with more accurate defect classification than the k-means clustering, allowing a wayside system with fewer sensors. Full article

Knowledge graph completion via link prediction is critical for intelligent equipment maintenance systems to support accurate fault diagnosis and maintenance decision making. However, existing approaches struggle to simultaneously capture local structural dependencies and perform effective multi-hop reasoning due to limited receptive fields or inefficient path exploration mechanisms. Traditional path-based methods implicitly assume path symmetry, treating all reasoning chains equally without considering their task-specific relevance. To address this issue, we propose a Graph Attention Network (GAT)-guided semantic path reasoning framework that breaks this symmetry through attention-driven asymmetric weighting, integrating local structural encoding with global multi-hop inference. The key innovation lies in a target-guided biased path sampling strategy, which transforms GAT attention weights into probabilistic transition biases, enabling adaptive exploration of high-quality semantic paths relevant to specific prediction targets. GATs learn importance-aware local representations, which guide biased random walks to efficiently sample task-relevant reasoning paths. The sampled paths are encoded and aggregated to form global semantic context representations, which are then fused with local embeddings through a gating mechanism for final link prediction. Experimental evaluations on FB15k-237, WN18RR, and a real-world equipment maintenance knowledge graph demonstrate that the proposed method consistently outperforms state-of-the-art baselines, achieving an MRR of 0.614 on the maintenance dataset and 0.485 on WN18RR. Further analysis shows that the learned path attention weights provide interpretable asymmetric reasoning evidence, enhancing transparency for safety-critical maintenance applications. Full article

The high-speed Electromagnetic Suspension (EMS) maglev guidance system exhibits inherent characteristics of strong nonlinearity, parameter time-variation, and complex external disturbances. To further optimize and improve the control performance of the guidance system for high-speed maglev trains, a novel intelligent control strategy that integrates the Deep Deterministic Policy Gradient (DDPG) algorithm with Super-Twisting Sliding Mode Control (STSMC) is proposed. Focusing on a single-ended guidance unit with differential control of dual electromagnets, an STSMC controller is first designed based on a cascaded control framework. To overcome the limitation of offline parameter tuning in dynamic operational conditions, a reinforcement learning optimization framework employing DDPG is introduced. A multi-objective hybrid reward function is formulated, incorporating error convergence, sliding mode stability, and chattering suppression, thereby realizing the online self-tuning of core STSMC parameters via real-time interaction between the agent and the environment. Numerical simulations under typical disturbance conditions verify that the proposed DDPG-STSMC controller significantly reduces the amplitude of guidance gap variation and accelerates dynamic recovery compared to conventional PID control. Its superior performance in disturbance rejection, control accuracy, and operational adaptability is validated. This study, conducted through high-fidelity numerical simulations based on actual system parameters, provides a robust theoretical foundation for subsequent hardware-in-the-loop (HIL) experimentation. Full article

This paper addresses the consensus problem in multi-agent systems via a self-learning control scheme that directly reuses prior control information to accelerate transient coordination while maintaining robustness. I study agents with linear dynamics and external disturbances, and design a lightweight self-learning consensus control law for the distributed consensus domain, formulated as $u_i\left(t\right)=k_1{u}_i(t-\tau )+k_2{s}_i\left(t\right)$ with learning intensity $k_1$ and learning interval $\tau$. I provide a Lyapunov-based stability proof showing uniform ultimate boundedness of the consensus error under bounded disturbances. Compared to non-learning consensus laws, the proposed strategy achieves faster agreement with reduced long-term effort and retains simplicity suitable for resource-constrained multi-agent platforms, while also achieving decent performance against external disturbances. Simulations validate the improved transient speed and steady accuracy. The full-version-source code is open-sourced. Full article

With the intensification of climate change, deepening energy transition, and increasing social vulnerability, extreme power outage events pose escalating challenges to the governance capacity of modern power systems. Existing evaluation frameworks primarily focus on engineering reliability and economic loss estimation, lacking systematic quantification of the governance complexity arising from multidimensional interacting pressures behind outage events. This creates a blind spot in both theoretical research and governance practice, hindering differentiated resilience decision-making. To address this gap, this study develops a four-dimensional evaluation framework of power outage governance complexity encompassing event attributes, external environment, internal system, and social impacts. Based on county-level outage data and multi-source auxiliary data in the United States from 2015 to 2024 and employing the XGBoost–SHAP interpretable machine learning approach, we construct the Power Outage Complexity Index (POCI) for all U.S. counties and systematically analyze its spatiotemporal evolution and core driving factors. The results show that outage governance complexity in the U.S. power grid exhibits a significant upward trend during 2015–2024, with an average annual growth rate of 1.84%. Spatially, significant positive autocorrelation is observed, and 146 high-complexity hotspot counties are identified, mainly clustered along the East and West Coasts, the Gulf Coast, and the Southwest. Driver analysis reveals that social impact and event attribute dimensions together account for nearly 90% of the variance in complexity, with cumulative outage exposure burden, outage frequency, and large-scale event ratio being the most critical drivers. Theoretically, this study extends power resilience research from an engineering-physical paradigm to a socio-technical governance paradigm and provides a reproducible methodological framework for assessing governance complexity in critical infrastructure systems. Practically, the POCI can serve as a governance diagnostic tool for the power industry and regulators, supporting resilience investment prioritization, emergency resource optimization, and differentiated governance strategy formulation. It also provides empirical evidence for safeguarding energy security in highly vulnerable communities and promoting energy resilience equity. Full article

The picking-head drivetrain is a critical transmission component of cotton harvesters, and its fault condition monitoring and diagnosis are essential for ensuring stable and reliable operation of the equipment. In practical engineering applications, diagnostic models for picking-head drivetrains are typically initialized using data collected under a limited number of representative operating conditions. Although sufficient fault samples can often be obtained during the initial training stage, the coverage of operating conditions is inherently restricted. As the model is deployed and used in the field, fault samples collected under new operating conditions are gradually acquired in a stage-wise manner. How to stably update the diagnostic model while the operating-condition coverage continuously expands, and how to avoid performance degradation and catastrophic forgetting, remain critical challenges. To address these issues, this paper proposes a continual learning method, termed DP-CL (Disentangled Prototype-Driven Continual Learning), for fault diagnosis of cotton harvester picking-head drivetrains under gradually expanding operating conditions. The proposed method is built upon an explicit disentanglement of condition-invariant features and condition-specific features. Within a unified framework, three types of structured prototypes, including class prototypes, condition prototypes, and condition-aware class prototypes, are constructed to form a multi-level representation hierarchy. A prototype-driven structured update mechanism is then employed to impose stable constraints on fault-discriminative semantics across different operating conditions. In addition, an operating-condition similarity measurement based on condition-specific features is introduced, based on which a proportion-adaptive sample selection strategy is designed. This strategy enables controlled knowledge transfer and preservation of discriminative structures during multi-stage model updates. Experimental results obtained under a laboratory-constructed cumulative operating-condition expansion scenario demonstrate that the proposed method achieves superior performance in terms of overall performance retention, cross-stage stability, and resistance to performance degradation. Moreover, as the number of operating conditions increases, the proposed method maintains a relatively smooth performance variation trend, while preserving clear class structures and a controllable level of confusion. These results validate the effectiveness of the proposed approach for stable fault diagnosis under expanding operating-condition coverage. Full article

Mn-based layered oxide cathodes are pivotal for advancing sodium-ion batteries, yet their practical deployment is hindered by structural instability and complex phase transformations during cycling. This review provides a systematic overview of recent strategies aimed at rational design and performance enhancement of these materials. It begins with fundamental thermodynamic principles governing phase formation, particularly P2/O3 structural dichotomy, and highlights the critical roles of sodium content, transition metal chemistry, and ionic potential in determining crystal stability. The emergence of high-entropy engineering is examined as a powerful approach to suppress detrimental phase transitions through configurational entropy stabilization, lattice distortion, and synergistic multi-element interactions. Furthermore, the integration of machine learning with multidimensional descriptors including electronegativity-weighted entropy and cationic potential enables more accurate predictions of phase behavior in complex compositional spaces. The review also highlights the decisive influence of synthesis protocols, where precise control over calcination conditions, atmosphere, and local elemental distribution enables the formation of targeted phase architectures, such as P2/O3 intergrowth, which exhibit superior electrochemical robustness. Collectively, these advances illustrate a shift from empirical trial and error toward a theory-guided, data-informed framework for designing high-performance layered oxide cathodes. Full article

Knowledge distillation serves as a prevalent model compression strategy within scene text detection, enabling the transfer of learned representations from a high-capacity teacher architecture to a streamlined student counterpart. Building upon this concept, deep mutual learning alleviates dependence on the teacher model through interactive learning among student models. However, existing deep mutual learning networks inadequately address the complex redundant backgrounds and text feature distributions in scene text images, failing to effectively balance the trade-off between model performance and lightweight design. To address this issue, this paper proposes an improved self-mutual learning framework based on deep mutual learning. By employing a design that incorporates parallel multi-detection heads and interactive learning, the proposed approach simplifies the model training process while significantly improving detection accuracy. Specifically, the framework introduces a pruning mechanism that enables different detection heads to capture input features with varying degrees of sparsity. This not only reduces interference from redundant backgrounds but also leads to a more lightweight implementation. Moreover, varying feature sparsity among detection heads promotes more diverse knowledge exchange throughout mutual learning. This substantially boosts the distilled model’s resilience in intricate text environments. Comprehensive evaluations show that our approach achieves superior F-measure scores compared to leading knowledge distillation methods. Full article

Precise drug target discovery is pivotal to mitigating the escalating costs and high attrition rates that characterize pharmaceutical research and development. Given that traditional single-omics methods often fail to elucidate the systemic complexity of human diseases, deep learning (DL)-enabled multi-omics integration has emerged as a transformative frontier. This review systematically summarizes the advancements in DL-driven multi-omics integration for drug target discovery. First, the multi-omics data foundation and integration strategies are delineated, followed by an exploration of the DL architectures utilized for processing such data. Subsequently, the efficacy of DL-driven multi-omics integration is examined regarding the identification of novel disease drivers, prediction of synthetic lethality interactions, and prioritization of therapeutic targets. Finally, addressing persistent challenges related to data sparsity, model interpretability, and target druggability and validation hurdles, emerging opportunities driven by Generative AI, Large Multimodal Models (LMMs), Explainable AI (XAI), and multidimensional feasibility assessment frameworks are discussed in the context of advancing precision medicine. Full article

Accelerated industrialization has caused complex mixed toxicant pollution, where synergistic or antagonistic interactions render conventional detection methods inadequate. Herein, we develop an integrated framework by pioneering the integration of microbial electrochemical systems (MECs) with machine learning (ML) for quantifying formaldehyde, tetracycline, Ag⁺, and Cu²⁺ in multi-component, multi-ratio, and multi-concentration mixtures. MECs generated dynamic current–time (I–t) signals responsive to toxicant stress, though signal overlap from mixed toxicants hindered direct quantification. Guided by toxicokinetics and electrochemical mechanisms, we developed a novel mechanism-driven feature engineering strategy with exclusively original indicators, which extracted 22 multidimensional features capturing instantaneous characteristics, kinetic patterns, and microbial stress-adaptive responses to resolve signal ambiguity, and provided biologically meaningful, high-information feature inputs that effectively bridge electrochemical response signals and ML modeling. Comparative analysis of four ML models (SVM, KNN, PLS, and RF) showed RF outperformed others, achieving R² > 0.9 for all toxicants (formaldehyde: 0.959; tetracycline: 0.934; Ag⁺: 0.936; Cu²⁺: 0.957) with minimized MAE and RMSE. Microbial community analysis identified Geobacter anodireducens (71.5%, electroactive for heavy metals) and Comamonas testosteroni (12.9%, organic degrader) as key functional taxa, supported by KEGG enzyme abundance data. This work overcomes traditional MEC limitations via innovative feature engineering and pioneering ML integration, providing a rapid, low-cost, and high-accuracy tool for environmental mixed toxicant monitoring. Full article

To address the rapid population diversity loss and premature convergence of the Artificial Lemming Algorithm (ALA) in complex optimization problems, this paper proposes an Improved Artificial Lemming Algorithm (IALA) with multi-strategy enhancements inspired by lemming behavior. First, a non-uniform mutation operator and a nonlinear step-size strategy are introduced to strengthen local optima escape capability and optimization precision. Second, inspired by the foraging and positioning behavior of lemmings, a relative advantage learning strategy is designed to reduce dependence on the global best individual, further enhancing the algorithm’s exploration ability. Finally, a Q-learning-based adaptive mechanism is integrated to intelligently orchestrate five lemming-inspired behavioral modes through a nonlinear reward function, enabling adaptive switching among search patterns. Comparative experiments on the CEC2022 benchmark suite demonstrate that IALA achieves a Friedman mean rank of 1.25, ranking first with a significant margin. Compared with the original ALA and other six classical and state-of-the-art metaheuristic algorithms, and four recently proposed improved ALA variants (EALA, IALA_Tan, DMSALAs, and MsIALA), the Wilcoxon rank-sum test confirms that IALA is significantly outperformed in only 2 out of 120 pairwise comparisons, exhibiting remarkable advantages in optimization accuracy, convergence speed, and robustness. Ablation experiments validate the synergistic necessity of all three strategies, with the Q-learning adaptive mechanism identified as the most critical contributor. Exploration–exploitation balance analysis and search history visualization further confirm that IALA achieves a smooth adaptive transition from global exploration to local exploitation. Space complexity analysis reveals that the Q-table introduces only approximately 19.5 KB of fixed additional overhead, which becomes negligible for high-dimensional problems. Furthermore, IALA is successfully applied to the parameter tuning of underwater manipulator controllers, verifying its efficiency and reliability in real-world engineering applications. Full article

Air pollution remains a major public health and environmental challenge in China, driven by complex non-linear interactions among anthropogenic activities, meteorological conditions, and topographic features that go beyond simple linear relationships. This study presents a comprehensive spatio-temporal assessment of key air pollutants (CO, NO₂, SO₂, and PM_2.5) and their relationships with Total Column Ozone (TCO) across China’s provinces from 2019 to 2023. Multi-source high-resolution satellite data from Sentinel-5P/TROPOMI, the China High PM_2.5 dataset, MODIS, and ERA5-Land reanalysis were integrated. A tiered analytical framework was applied, combining linear Pearson correlations, non-linear Spearman rank correlations, and interpretable XGBoost machine learning with SHAP values. Results reveal a distinct seasonal “seesaw” pattern, with primary pollutants peaking during winter stagnation and TCO reaching maximum levels in late winter and spring. Non-linear analyses uncover critical threshold effects, including exponential increases in PM_2.5 and SO₂ when surface temperatures drop below 0 °C, very strong SO₂-TCO coupling (ρ = 0.93), and significant pollutant trapping in low-elevation regions (CO-elevation ρ = −0.82). These findings support the development of precision environmental policies with dynamic, temperature-threshold-based emission controls and topography-specific strategies to effectively mitigate air pollution in China. Full article

The pursuit of energy efficiency and sustainability in the built environment has placed façade systems at the forefront of innovation in architectural design. This study proposes an integrated framework that combines generative design techniques with artificial intelligence (AI) to optimize composite façade configurations for next-generation smart buildings. Using parametric modeling, a wide design space of façade geometries and material compositions was generated, capturing trade-offs between thermal performance, daylight, structural strength, and aesthetic variability. Artificial intelligence algorithms, particularly machine learning models, are trained on simulation-derived performance datasets to rapidly predict key indicators such as energy consumption, thermal transmittance (U-value) and solar heat gain coefficients. The proposed approach achieved a predictive accuracy of 99.85%, enabling efficient exploration of optimal solutions across high-dimensional design alternatives. A multi-objective optimization strategy was further implemented to balance energy efficiency with structural and aesthetic constraints, producing façade configurations that outperform conventional designs. The findings demonstrate that integrating generative design with AI-based prediction not only accelerates the façade design process but also provides actionable pathways toward net-zero energy buildings. This research highlights the transformative potential of AI-driven generative workflows in advancing sustainable architecture and delivering intelligent, adaptive and performance-oriented façades for future urban environments. Full article

The increasing penetration of inverter-interfaced photovoltaic (PV) generation in active distribution networks (ADNs) intensifies fast voltage violations and makes real-time Volt-VAR control (VVC) challenging, especially when each inverter has only partial and noisy measurements and communication is limited. Existing local droop-type strategies lack coordination, while fully centralized optimization/learning is often impractical for online deployment. To address these gaps, an attention-enhanced multi-agent deep reinforcement learning (MADRL) framework is developed for inverter-based VVC under the centralized training and decentralized execution (CTDE) paradigm. First, the voltage regulation problem is formulated as a decentralized partially observable Markov decision process (Dec-POMDP) to explicitly account for system stochasticity and temporal variability under partial observability. To solve this complex game, an attention-enhanced MADRL architecture is employed, where an agent-level attention mechanism is integrated into the centralized critic. Unlike traditional methods that treat all neighbor information equally, the proposed mechanism enables each inverter agent to dynamically prioritize and selectively focus on the most influential states from other agents, effectively capturing complex intercorrelations while enhancing training stability and learning efficiency. Operating under the CTDE paradigm, the framework realizes coordinated reactive power support using only local measurements, ensuring high scalability and practical implementability in communication-constrained environments. Simulations on the IEEE 33-bus system with six PV inverters show that the proposed method reduces the average voltage deviation on the test set from 0.0117 p.u. (droop control) and 0.0112 p.u. (MADDPG) to 0.0074 p.u., while maintaining millisecond-level execution time comparable to other MADRL baselines. Scalability tests with up to 12 agents further demonstrate robust performance of the proposed method under higher PV penetration. Full article