Search Results (545)

Background: Agentic artificial intelligence (AI) systems, characterized by autonomous goal-directed behavior, multi-step reasoning, task decomposition, and tool use, are increasingly proposed for healthcare applications. However, their autonomy raises concerns regarding transparency, accountability, and human oversight. While explainable AI (XAI) has been widely studied in traditional predictive models, less is known about how explainability is implemented within agentic architectures. Objective: To map the emerging literature on explainable agentic AI (XAAI) in healthcare and characterize the types, scope, and forms of explainability used in these systems. Methods: A scoping review was conducted following PRISMA-ScR guidelines. PubMed, Embase, IEEE Xplore, and ACM Digital Library were searched through November 2025. Eligible studies described healthcare-related agentic AI systems incorporating explicit explainability mechanisms. Data were extracted on system architecture, explainability type (intrinsic, post hoc, hybrid), explanation scope (local, global), explanation form, and reported clinical outcomes. Results: Nine studies met the inclusion criteria. All systems demonstrated core agentic features, including autonomy, task decomposition, and tool integration, often within multi-agent frameworks. Explainability was predominantly intrinsic and workflow-native, typically delivered through textual reasoning traces and example-based grounding in retrieved clinical evidence. Feature-based and global explanations were comparatively rare and largely confined to hybrid architectures. Across domains including radiology, neurology, psychiatry, and biomedical research, XAAI systems were reported to improve performance and interpretability relative to baseline models in the included studies. However, these findings were derived from heterogeneous, predominantly experimental or retrospective studies, and structured human-in-the-loop oversight was infrequently described. Conclusions: Current XAAI systems appear to emphasize process transparency and evidence grounding rather than mechanistic model-level attribution. The available evidence remains limited and heterogeneous, and findings should be interpreted as early trends rather than established characteristics. Further progress will require standardized evaluation frameworks, clearer reporting of oversight mechanisms, and validation in real-world clinical settings to support safe and trustworthy integration of agentic AI into healthcare practice. Full article

Accurate ultra-short-term photovoltaic (PV) power forecasting is essential for secure grid dispatch and renewable-rich system operation, yet it remains difficult because of rapid weather fluctuations and error accumulation in multi-step prediction. This paper proposes a decoupled physics-guided residual-correction framework, built on an attention-based sequence-to-sequence (Seq2Seq) architecture, for deterministic 4 h ahead rolling PV forecasting at 15 min resolution. In the first stage, a physical model maps numerical weather prediction (NWP) inputs to a deterministic baseline trajectory while preserving physical bounds. In the second stage, an Attention-Seq2Seq network learns the structured residual trajectory from historical sequences. The global attention mechanism allows the decoder to focus on the most informative historical states, helping reduce information loss and error accumulation over extended horizons. Experiments on a 22-month real-world PV dataset show that the proposed framework outperforms conventional linear and nonlinear benchmarks, reducing root mean square error (RMSE) and mean absolute error (MAE) by 23.79% and 39.17%, respectively, relative to the physical baseline. The framework also maintains robust instantaneous tracking under rapidly changing cloud conditions and preserves a 30–40% error reduction rate at Steps 12–16, supporting reliable intraday scheduling. Full article

Energy-intensive refrigeration systems require predictive models that remain informative under counterfactual control trajectories, not only on archived operation. This paper develops a control-oriented multi-step state model and a nonlinear model predictive control framework for an indoor ice-rink refrigeration system. Historical state, control, and exogenous variables are encoded jointly with an admissible future control trajectory, and a normalized thermal-balance residual is added to the training objective. A lightweight conditioned transformer predicts ice temperature, return-glycol temperature, supply-glycol temperature, and compressor power over a 30 min horizon. The selected weakly regularized model with regularization coefficient λ_phys = 0.001 decreases the normalized thermal-balance root-mean-square error on the horizon tail by 30.29% relative to the base model while increasing the average ice-temperature root-mean-square error by only 1.90%. In a surrogate-based counterfactual four-day evaluation, the resulting nonlinear model predictive controller reduces predicted daily energy by 4.84%, terminal violation share by 17.32%, mean absolute terminal ice-temperature deviation by 18.74%, and the mean objective value by 30.82% relative to historical admissible setpoint tracking. The mean full control cycle time is 0.0311 s, confirming real-time feasibility for a 5 min supervisory update interval. All controller results are surrogate-based rather than field-deployed and therefore represent receding-horizon benchmark results under learned-model evaluation, not realized field savings. Full article

To address the limited accuracy of power demand prediction for hybrid electric vehicles under complex and dynamic driving conditions, this paper proposes a hybrid prediction approach based on the cascade correction of Long Short-Term Memory networks and Least Squares Support Vector Machines (LSTM-LSSVM-CC). The proposed method adopts a stage-wise modeling framework that exploits the least-squares optimality of LSSVM for low-frequency steady-state signals and the dynamic compensation capability of LSTM for high-frequency non-stationary residuals, thereby achieving complementary feature representation in the frequency domain. Specifically, an LSSVM is first used to construct a baseline regression model that captures stationary components, followed by an LSTM network that performs deep temporal modeling of the residual sequence to correct nonlinear prediction errors. Extensive experiments conducted on three standard driving cycles—CLTC-P, WLTP, and UDDS—demonstrate that the proposed model consistently outperforms conventional methods including LSSVM, RNN, ELMAN, and Random Forest in multi-step predictions, achieving an average RMSE reduction of 28–52% and maintaining correlation coefficients (R²) between 0.87 and 0.99. Particularly under highly dynamic and abrupt load conditions, the model exhibits superior real-time performance and stability while significantly mitigating cumulative prediction errors. These results demonstrate that the proposed LSTM-LSSVM-CC model achieves robust modeling performance of non-stationary time series while balancing prediction accuracy and computational efficiency, providing an effective technical foundation for hybrid vehicle energy management optimization and offering a transferable theoretical framework for time-series prediction in complex systems. Full article

Existing studies of street-related emotional perception mainly rely on static scene evaluations, which cannot capture the cumulative effects of environmental exposure during continuous walking. To address this limitation, this study proposes a method for predicting pedestrian physiological responses in sequential audiovisual street environments. Four real-world walking routes were selected, with outbound and return directions treated as independent paths, yielding eight paths and 32 valid samples. EEG, ECG, sound pressure level, first-person video, and GPS data were synchronously collected to construct a 1 s multimodal time-series dataset. Pearson correlation, Kendall correlation, and mutual information analyses were used to examine linear, monotonic, and nonlinear relationships between environmental variables and physiological indicators, and the resulting weights were incorporated into a Long Short-Term Memory (LSTM) model for multi-step prediction. Visual elements and noise exposure were the main factors influencing physiological responses. Among the models, the mutual-information-weighted LSTM performed best, achieving an R² of 0.77 for heart rate variability (RMSSD), whereas prediction of the EEG ratio (β/α and θ/β) remained limited. An additional independent street sample outside the training set was then used to generate a dual-dimensional EEG-ECG physiological response map, demonstrating the model’s potential for identifying emotional risk segments and supporting street-level micro-renewal. Full article

Background/Objectives: Population-specific genomic data are essential for understanding hepatocellular carcinoma (HCC) biology, particularly in underrepresented regions. This study aimed to perform exploratory single-nucleotide polymorphism (SNP)-array-based profiling of HCC tumor samples from Indonesian patients and to prioritize candidate functional variants using a systematic in silico framework. Methods: This retrospective cross-sectional study included 15 resected HCC cases with available formalin-fixed paraffin-embedded (FFPE) tumor tissue. Genome-wide SNP genotyping was performed using the Illumina Asian Screening Array. Following quality control and filtering, variants were annotated using the Ensembl Variant Effect Predictor. A case-only functional prioritization approach incorporating multiple in silico prediction tools was applied, followed by gene-level burden aggregation. Results: After multistep filtering, 11 samples and 104 prioritized variants were retained for analysis. Variants consisted predominantly of splice-region, missense, and regulatory changes. Gene-level burden analysis identified Nuclear Autoantigenic Sperm Protein (NASP, rs775916096) as the highest-ranked candidate gene, while G protein-coupled receptor 78 (GPR78, rs558447540) emerged as a secondary candidate with predicted functional annotations but currently limited biological evidence in HCC. Given the tumor-only design without matched normal tissue, the prioritized variants cannot be distinguished from rare germline variants. Conclusions: This exploratory SNP-array study provides a hypothesis-generating framework for functional variant prioritization in Indonesian HCC. NASP and GPR78 represent preliminary candidates that require validation in larger cohorts with matched normal tissue and sequencing-based confirmation. Full article

Accurate representation of short-term reservoir water-level dynamics is essential for operational analysis and scenario-based assessment under prescribed inflow–outflow conditions. In many practical applications, physically based modelling is limited by incomplete process knowledge, unavailable boundary conditions, or insufficient temporal resolution of input data. This study presents a data-driven framework for hourly conditional simulation of reservoir water level based on a hybrid Conv1D–LSTM architecture. The model learns nonlinear relationships among hydraulic forcing, operational control, and system state from historical observations, and is evaluated in a recursive multi-step simulation (rollout) mode to reflect its intended use and capture error accumulation over time. A systematic analysis of input sequence length and activation function is performed to identify a robust model configuration. On the test set, the selected configuration ($L = 24$, GELU) achieved RMSE = 0.1057 m, MAE = 0.0881 m, and R² = 0.972 in rollout evaluation. The proposed framework is designed for scenario-based simulation rather than one-step deterministic forecasting, enabling rapid operational screening of alternative inflow–outflow regimes. Unlike many previous studies that emphasize one-step predictive accuracy, this work explicitly assesses model stability in recursive multi-step simulation, which is more relevant for reservoir scenario analysis. Full article

Flexible deformation sensors inevitably suffer from sensitivity degradation and severe measurement errors during long-term cyclic stretching due to structural fatigue. Traditional material-level optimizations are costly and lack dynamic adaptability. Herein, we propose an artificial intelligence (AI)-driven predict-and-compensate framework for the online calibration of flexible sensors. To overcome training sample scarcity, a generative adversarial network (GAN) performs temporal data augmentation. Subsequently, a hybrid deep learning framework integrating long short-term memory (LSTM) networks and a Sequence Attention mechanism is employed. This architecture accurately captures both local signal fluctuations and multiscale long-term decay trends, enabling precise multi-step prediction and output compensation. Experimental evaluations validate that this strategy significantly suppresses sensor response drift. Under cyclic loading, an initially substantial relative measurement error of 48.63% plummets to 7.16% post-calibration, with typical errors consistently reduced to the ~1% level. Furthermore, when deployed in a smart glove gesture recognition system, this method successfully restores the recognition accuracy from a fatigue-induced low of 75.73% (after 200 stretch cycles) back to 97.70%. This generative and attention-based deep learning paradigm offers robust, real-time error calibration, providing a highly viable solution for extending the long-term reliability and stability of flexible sensor systems. Full article

Karst aquifers present unique challenges for groundwater level prediction due to their dual-porosity structures and highly nonlinear hydrological responses. This study systematically evaluates nine machine learning and deep learning models (RF, XGBoost, LSTM, CNN, Transformer, N-BEATS, CNN-LSTM, Seq2Seq-LSTM, and Attention-Seq2Seq-LSTM) for rainfall-driven groundwater level forecasting in the Maocun subterranean river catchment, Guilin, Guangxi, China. Two years of hourly high-frequency data from three monitoring sites representing distinct hydrogeological zones (recharge, flow, and discharge) were employed within a multidimensional evaluation framework integrating single-step accuracy, multi-step stability, and computational efficiency. Results indicate that the Transformer achieved the highest single-step prediction accuracy, attaining the lowest RMSE (0.130–0.606 m) and highest R² (0.813–0.965) across all three sites. CNN-LSTM offered the best balance between predictive performance and computational cost, requiring an average training time of only 27.97 s and 28.0 convergence epochs. N-BEATS demonstrated superior long-term stability in 12-steps-ahead forecasting, achieving R² = 0.914 at ZK1, outperforming all other architectures. More fundamentally, hydrogeological complexity exerted a dominant control on predictive skill that systematically outweighed differences arising from model architecture. All models yielded R² below 0.813 at the geologically complex ZK2 site, whereas R² exceeded 0.950 across all models at ZK1, indicating that aquifer complexity, rather than algorithm selection, constitutes the primary constraint on prediction feasibility. This study presents the first application of N-BEATS to karst groundwater level forecasting and proposes a replicable multidimensional evaluation framework, providing a scientific reference for intelligent modelling of complex karst systems. Full article

With the rapid development of new energy vehicles and energy storage systems, accurate prediction of the remaining useful life (RUL) of lithium-ion batteries is of great importance for predictive maintenance and operational safety. However, battery degradation during cycling usually exhibits multi-scale characteristics, including long-term degradation trends, stage-wise drifts, and stochastic disturbances, which makes existing methods still face significant challenges in multi-step forecasting and cross-domain generalization. To address this issue, this paper proposes a time–frequency fusion model for multi-step RUL prediction, termed TF-RULNet (Time-Frequency RUL Network). The model takes cycle-level feature sequences as input and consists of three components: a multi-scale temporal convolution encoder (MSTC) for parallel extraction of degradation cues at different temporal scales; a multi-head spectral interaction module (MHSI), which performs 1D-FFT along the temporal dimension for each head and further applies adaptive band-wise mask refinement to capture local spectral structures and hierarchical band patterns with a computational complexity of $O(LlogL)$; and a cross-gated fusion module (CGF), which generates gating signals from the summary of one domain to modulate the features of the other domain, thereby enabling dynamic balancing and complementary enhancement of time–frequency information. Experiments are conducted on the NASA dataset (B005/B007) for in-domain evaluation, and further cross-dataset tests from NASA to the Maryland dataset (CS-35/CS-37) are carried out to verify the robustness of the proposed model under distribution shifts. The results show that, compared with the strongest baseline PatchTST, TF-RULNet reduces RMSE and MAE by more than 38.23% and 50.51%, respectively, in cross-dataset generalization, while achieving an additional RMSE reduction of about 24% in in-domain prediction. In summary, TF-RULNet can effectively characterize the multi-scale time–frequency degradation patterns of batteries and improve cross-domain generalization, providing a high-accuracy and scalable modeling solution for practical battery health management and life prognostics. Full article

Background/Objectives: Oral squamous cell carcinoma (OSCC) commonly arises from oral potentially malignant disorders (OPMDs), yet reliable molecular biomarkers that predict malignant transformation remain scarce. Because epithelial carcinogenesis follows similar multistep trajectories across multiple organs, pan-cancer transcriptional analyses may reveal conserved pathways relevant to early oral tumorigenesis. This study aimed to identify shared transcriptional signatures across carcinomas and evaluate their applicability to precancerous-to-carcinoma progression. Methods: Bulk RNA-seq data from five carcinomas (lung, colon, breast, prostate, and head and neck squamous cell carcinoma, HNSCC) were obtained from TCGA to identify shared differentially expressed genes (DEGs) (|log₂FC| ≥ 2; FDR < 0.05). Functional enrichment, clustering, and gene–pathway network analyses characterized conserved biological processes. Independent GEO datasets containing premalignant and malignant samples, including OPMD and OSCC cohorts, were examined to assess early-stage relevance. Results: A conserved 45-gene signature was identified, enriched for transcriptional regulation, chromatin organization, and RNA polymerase II-mediated processes. Regulatory hubs, including ZIC5, MYBL2, ONECUT2, POU4F1, and PDX1, and strong upregulation of cancer-testis antigens (MAGEA3, MAGEA6, MAGEC2) were notable. Integration with premalignant datasets revealed 13 genes consistently dysregulated across early lesions, involving pathways such as cell differentiation, apoptosis, and lipid transport. Several genes remained altered from normal tissue through OPMD to OSCC, supporting their potential as stable biomarkers. Conclusions: This study identifies conserved transcriptional programs shared across epithelial cancers and detectable in OPMDs. These findings highlight promising biomarker and regulatory candidates for improving early detection and risk stratification of oral precancer, addressing a critical unmet need in OSCC prevention and clinical management. Full article

Cloud cover significantly reduces the electrical power output of photovoltaic systems, making accurate short-term cloud movement predictions essential for reliable solar energy production planning. This article presents a deep learning framework that directly estimates cloud movement from ground-based all-sky camera images, rather than predicting future production from past power data. The system is based on a three-step process: First, a lightweight Convolutional Neural Network segments cloud regions and produces probabilistic masks that represent the spatial distribution of clouds in a compact and computationally efficient manner. This allows subsequent models to focus on the geometry of clouds rather than irrelevant visual features such as illumination changes. Second, a Vector Quantized Variational Autoencoder compresses these masks into discrete latent token sequences, reducing dimensionality while preserving fundamental cloud structure patterns. Third, a GPT-style autoregressive transformer learns temporal dependencies in this token space and predicts future sequences based on past observations, enabling iterative multi-step predictions, where each prediction serves as the input for subsequent time steps. Our evaluations show an average intersection-over-union ratio of 0.92 and a pixel accuracy of 0.96 for single-step (5 s ahead) predictions, while performance smoothly decreases to an intersection-over-union ratio of 0.65 and an accuracy of 0.80 in 10 min autoregressive propagation. The framework also provides prediction uncertainty estimates through token-level entropy measurement, which shows positive correlation with prediction error and serves as a confidence indicator for downstream decision-making in solar energy forecasting applications. Full article

In this study, a novel direct torque control (DTC) strategy is proposed to mitigate the torque ripple issue inherent in switched reluctance motors (SRMs), which is caused by the double salient pole configuration and the pulse power supply mode. The strategy is based on the prediction and optimization of a long-time-domain model. Central to this method is the development of a multi-step predictive optimization framework. By incorporating hysteresis control, the conventional approach of minimizing instantaneous error in predictive control is shifted towards minimizing tracking error over an extended time frame. A dual-objective evaluation function is also introduced, which simultaneously optimizes both torque smoothness and switching frequency, ensuring their collaborative enhancement. To validate the proposed method, a 6/4-pole SRM simulation model was implemented using MATLAB/Simulink 2024B, and comparisons were made with traditional methods. The results demonstrate that this strategy significantly reduces torque pulsation and lowers the system’s switching frequency, even under varying operational conditions such as different rotational speeds and sudden load variations. Consequently, this approach not only guarantees improved dynamic performance but also enhances the motor’s efficiency and stability. Full article

Short-term load forecasting plays a pivotal role in modern power system operations yet it remains challenging due to the complex spatiotemporal dependencies in load data. This paper proposes a dual-head attention residual network (DARNet) that significantly advances STLF through three key innovations: (1) a hybrid encoder combining 1D-CNN and GRU architectures to simultaneously capture the local load patterns and long-term temporal dependencies, achieving a 28% better locality awareness than that of conventional approaches; (2) a novel dual-head attention mechanism that dynamically models both the inter-temporal relationships and cross-variable dependencies, reducing the feature engineering requirements; and (3) an autocorrelation-adjusted recursive forecasting framework that cuts the multi-step prediction error accumulation by 33% compared to that with standard seq2seq models. Extensive experiments on real-world datasets from three Chinese cities demonstrate DARNet’s superior performance, outperforming six state-of-the-art benchmarks by 21–35% across all of the evaluation metrics (MAPE, SMAPE, MAE, and RRSE) while maintaining robust generalization across different geographical regions and prediction horizons. Full article

Accurate monitoring of crop phenology is essential for precision agriculture and yield forecasting. However, satellite-derived time series often suffer from inherent noise, such as residual atmospheric effects and mixed pixels, as well as a frequent lack of ground-truth data in agriculture. In response, this study proposes an algorithm to define the type of spectral signatures for the principal phenological stages of crops, using them as the foundation for training supervised machine learning classification models. The algorithm was developed using Fuzzy C-Means (FCM) clustering to identify the spectral signature reference groups in winter wheat across the Burgos region (Spain) during the 2020 and 2021 growing seasons. To enhance cluster independence and biological coherence, a multi-step filtering process was implemented, including spectral purity (membership degree, SAM, and SAMder) and temporal coherence filters. The filtered and labeled dataset (80% original Burgos dataset) was used to train supervised classification models (KNN and XGBoost). The models’ reliability was verified through three wheat tests (remaining 20%), labeled using other clustering techniques, and an independent barley dataset from diverse geographic locations (Valladolid and Soria). The filtering process significantly improved cluster stability by removing outliers and transition spectral signatures. The supervised models demonstrated exceptional performance; the KNN model slightly outperformed XGB, achieving a mean Accuracy of 0.977, a Kappa of 0.967, and an F1-score of 0.977 in the wheat external test. Furthermore, the model showed, when applied to barley, that its phenological spectral signatures are equivalent in shape to those of wheat, with an Accuracy of 0.965 and an F1-score of 0.974. In addition, it was verified that the type spectral signatures remain the same regardless of the location. This study presents a robust classification tool capable of labeling four key phenological stages (tillering, stem elongation, ripening, and senescence) without ground truth. By effectively removing inherent satellite noise, the proposed methodology produces organized, cleaned datasets. This structured foundation is critical for future research integrating spectral signatures with harvester data to develop high-precision yield prediction models. Full article