Search Results (5,094)

Deep learning algorithms have become useful tools for predicting time-series displacements from historical displacements measured using interferometric synthetic aperture radar (InSAR) techniques. However, nearly all existing InSAR-related studies are based on data-driven deep learning algorithms, causing poor robustness, especially for long-term prediction with small-scale training samples. In this study, we propose a new algorithm, named physics-informed neural network with prior parameter inversion (PINNPI), for InSAR-based prediction of time-series displacements. PINNPI is a hybrid data-driven and knowledge-guided deep learning network, where two coupled deep neural networks are first constructed for network training and parameter inversion of prior knowledge. The outputs of these two deep neural networks are coupled by an automatic differentiation module. By minimizing a hybrid physics-informed and data-driven loss function, the proposed network simultaneously models time-series displacement and estimates prior parameters. Subsequently, time-series displacements are predicted based on the trained networks and inverted parameters. The incorporation of physical knowledge into PINNPI enhances the capability of long-term displacement prediction with respect to data-driven learning algorithms. In addition, PINNPI effectively improves the poor robustness of classical PINNs, when prior parameters are unknown. Simulations and two real-world tests suggest that the accuracy of displacement prediction by PINNPI is, on average, 85% and 88% higher than that of classical data-driven deep learning and PINN algorithms, respectively. This work offers a new insight for predicting InSAR-based displacements associated with anthropogenic and geophysical activities. Full article

Reliable preconstruction cost estimation (PCE) is a fundamental stage to project planning and investment decision-making. However, the early-stage uncertainty and limited information make early-stage cost estimation challenging. Artificial intelligence has attracted growing interest as a way out of this impasse, producing a substantial empirical literature worth systematic examination. This study synthesizes the findings of 30 empirical studies published up to December 2025 using PRISMA protocols and examines the AI techniques, project types, dataset characteristics, validation practices, and model interpretability. The results show that artificial neural networks (ANNs) and hybrid approaches dominate the literature, with applications concentrated in building and transportation projects. Reported model performance is generally strong across commonly used evaluation metrics. However, several structural limitations persist: poor generalizability across project contexts, inconsistent validation procedures, limited adoption of explainable AI, and minimal integration of domain expertise. These factors, together, limit the transferability of existing models to real-world practice. This review contributes a structured methodological synthesis, maps the gaps that most limit progress, and proposes a conceptual AI–Expert Integration Framework to support estimation approaches that are more robust, interpretable, and decision-oriented. The findings offer both a current assessment of the field and a practical roadmap for advancing AI-driven PCE research. Full article

Hybrid quantum–classical machine learning offers a promising direction for advancing automated quality control in industrial settings. In this study, we investigate two hybrid quantum–classical approaches for classifying defects in aluminum TIG welding images and benchmarking their performance against a conventional deep learning model. A convolutional neural network is used to extract compact and informative feature vectors from weld images, effectively reducing the higher-dimensional pixel space to a lower-dimensional feature space. Our first quantum approach encodes these features into quantum states using a parameterized quantum feature map composed of rotation and entangling gates. We compute a quantum kernel matrix from the inner products of these states, defining a linear system in a higher-dimensional Hilbert space corresponding to the support vector machine (SVM) optimization problem and solving it using a Variational Quantum Linear Solver (VQLS). We also examine the effect of the quantum kernel condition number on classification performance. In our second method, we apply angle encoding to the extracted features in a variational quantum circuit and use a classical optimizer for model training. Both quantum models are tested on binary and multiclass classification tasks, and the performance is compared with the classical CNN model. Our results show that while the CNN model demonstrates robust performance, hybrid quantum–classical models perform competitively. This highlights the potential of hybrid quantum–classical approaches for near-term real-world applications in industrial defect detection and quality assurance. Full article

Membrane technologies play a role in water treatment, energy conversion, and industrial separation processes; however, their performance is limited by fouling, polarization phenomena, transport inefficiencies, and energy consumption. This study presents a review of the integration of computational fluid dynamics and machine learning in membrane technologies, complemented by an environmental and engineering assessment of representative membrane materials. A systematic literature screening based on PRISMA guidelines was conducted using the Scopus (Elsevier B.V., Amsterdam, The Netherlands) and Web of Science (Clarivate, Philadelphia, PA, USA) databases, yielding 1421 records, of which 54 studies met the predefined relevance criteria. The analysis revealed a transition from conventional physics-based approaches toward hybrid simulation–machine learning frameworks, with artificial neural networks, surrogate models, and optimization emerging as the dominant methodologies. Energy consumption was identified as the most frequently investigated variable, particularly in desalination, fuel cell, electrodialysis, and hydrogen production systems. A complementary material-level assessment showed that conventional polymeric membranes, especially polyamide-based systems, remain dominant due to their performance and economic feasibility, whereas advanced materials such as graphene, carbon nanotubes, and perovskites offer promising functional properties but face challenges. The findings highlight the potential of integrated simulation–machine learning–material assessment frameworks to accelerate the development of intelligent and sustainable membrane technologies for future applications. Full article

This survey presents a large-scale, reproducible, and parameter-mapped benchmarking analysis of path planning algorithms for multiple mobile robot systems (MMRS) by systematically examining 247 rigorously filtered papers from high-impact journals. Unlike prior reviews that primarily provide conceptual taxonomies, this survey introduces execution-oriented multi-parameter mapping enabling direct comparison of classical planners (A*, D*, Cell Decomposition, APF, RM, RRT, and ORCA), nature-inspired metaheuristics (PSO, GA, ACO, GWO, FA, ABC, BFO, CS, BA, SFLA, eagle-inspired optimizers), and learning-driven AI frameworks (Fuzzy Logic, Artificial Neural Networks, and Deep Reinforcement Learning). Each paper is evaluated across 15 practical planning dimensions, including environment type (static 95% vs. dynamic 51%), multi-robot validation (52%), dynamic goal handling (13%), energy awareness (14%), timepath optimization bias (82% focus), inter-robot coordination (less than 47%), and software validation platforms (MATLAB 42% and ROS 9%), revealing that simulation-only validation dominates (98%) while experimental testing remains limited (33%). Multivariate validation through Multiple Correspondence Analysis further confirms that coordination maturity, energy awareness, and multi-robot applicability are the primary structural differentiators of deployment readiness across algorithm families. The findings emphasize the need for hybrid, energy-aware, and coordination-driven MRPP frameworks supported by experimental benchmarking and reproducible deployment pipelines to advance real-world MMRS autonomy. Full article

Background: Robotic rehabilitation systems for upper-limb stroke rehabilitation have been developed across diverse robotic platforms, yet cross-study comparisons remain challenging due to heterogeneous system designs and classification approaches. This brief narrative review proposes a paradigm-driven framework, categorizing robotic rehabilitation systems based on underlying therapeutic interaction principles (e.g., mirror, bimanual, mirror–bimanual hybrid) rather than implementation modality alone. Methods: A structured MEDLINE/PubMed literature search was performed to identify representative studies describing robotic system characteristics, rehabilitation task structures, clinical outcomes, and mechanistic insights within mirror, bimanual, and hybrid rehabilitation paradigms. Findings: Among the reviewed studies, mirror-based systems emphasize sensory representation and interhemispheric modulation, whereas bimanual systems target coordination and motor learning through bilateral interaction. Hybrid systems integrate these approaches by combining mirrored feedback with active bilateral engagement. Emerging neuroimaging evidence, particularly resting-state fMRI, may help relate clinical improvements to neural network changes, providing a mechanism-informed perspective on rehabilitation outcomes. Conclusions: This review highlights substantial progress in robotic upper-limb stroke rehabilitation, with current systems increasingly integrating multimodal feedback and task-oriented interaction within mirror, bimanual, and hybrid rehabilitation paradigms. Limitations include variability across studies due to differences in robotic system implementation, task design, duration, stroke chronicity, and patient engagement. Full article

Driver behavior is a critical factor in road safety, contributing to the majority of traffic crashes. The i-DREAMS project introduced the concept of a Safety Tolerance Zone (STZ) to enhance driving safety through real-time and post-trip interventions. This study develops and evaluates three hybrid machine learning models—(i) Deep Neural Network–Random Forest (DNN-RF), (ii) Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM), and (iii) Recurrent Neural Network–AdaBoost (RNN-AdaBoost)—to classify risky driving behavior into three safety levels using naturalistic driving data from Belgium and the UK. The dataset includes 69 drivers, 15,389 trips, and 265,512 min of driving data. Among the models tested, the DNN-RF model demonstrated the highest accuracy, reaching 98% in Belgium and 97% in the United Kingdom, outperforming other approaches. Feature importance analysis identified harsh acceleration and braking as the most critical factors in Belgium, while total trip distance and harsh acceleration were predominant in the UK. To enhance model transparency, we applied the Local Interpretable Model-agnostic Explanations (LIME) algorithm, providing valuable insights into model predictions. The findings support the potential of hybrid deep learning models in improving road safety by accurately detecting risky driving behaviors. These insights can inform targeted interventions and driver assistance technologies to mitigate crash risks and promote safer driving practices. Full article

Assessing earthquake-induced landslide (EQIL) susceptibility is essential for hazard mitigation in mountainous regions. While background hydrological variations influence slope stability, long-term mean soil moisture is rarely incorporated into deep learning-based landslide susceptibility mapping (LSM). This study proposes a hybrid Convolutional Neural Network and Swin Transformer (CNN-SwinT) framework that integrates long-term mean soil moisture as a static covariate to represent persistent background moisture conditions. The model couples the local spatial feature extraction of CNNs with the hierarchical contextual representation of Swin Transformers to capture multi-scale spatial dependencies. Using Minxian County of China as the study area, thirteen conditioning factors were selected via multicollinearity and information gain ratio analyses. The dataset was split into training (70%) and validation (30%) sets. Performance comparison against standalone CNN and SwinT models revealed that the hybrid CNN-SwinT achieved the highest accuracy (0.856) and AUC (0.95), with predicted high-susceptibility zones closely aligning with historical inventories. However, these reported metrics reflect a random, spatially non-independent split, and spatial block cross-validation is recommended for future operational deployment. The results demonstrate that incorporating long-term mean soil moisture provides critical complementary hydrological information that enhances predictive performance. These findings indicate that the proposed hybrid framework is reliable and effective for high-resolution EQIL susceptibility mapping. Full article

In a highly homogeneous market environment, consumers’ cognitive inertia towards standardized products continues to strengthen, and traditional functional attributes significantly weaken the driving force of impulse buying. This study is based on the elaboration likelihood model (ELM). It constructs a dual-path theoretical framework of “cognitive inertia cue arousal”, aiming to reveal a new triggering mechanism for impulse buying behavior (IBB) in homogeneous competition. This study proposes a hybrid structural equation model (SEM) and artificial neural network (ANN) approach to evaluate the driving factors behind consumers’ maintenance of IBB in the context of homogeneous competition. Specifically, the first stage involves identifying key variables that significantly affect IBB using SEM. The second stage consists of predicting the accuracy of the outcome variables and ranking their importance using an ANN. This study confirms two pathways: product cognitive inertia (customized information and promotional discounts) and peripheral cue arousal (anchor novelty display strategy and observation learning). Research has found that observational learning has a positive impact on perceived attention. Interestingly, perceived attention does not have a significant mediating effect between promotional discounts and IBB. This discovery provides a theoretical basis and practical paradigm for cracking the involution competition. Full article

Background/Objectives: Alzheimer’s disease (AD) is a progressive neurodegenerative disorder necessitating accurate and timely diagnosis for effective clinical intervention. While deep learning methods have shown promise in AD classification, many rely on computationally intensive architectures and high-dimensional feature representations. This study introduces a lightweight hybrid framework combining deep feature extraction, dimensionality reduction, and adaptive classification for MRI-based Alzheimer’s disease stage classification. Methods: Utilizing MRI images from a publicly available Alzheimer’s disease dataset encompassing four clinical stages (Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented), deep features were extracted using a pre-trained SqueezeNet model as a fixed feature extractor, generating 1000-dimensional feature vectors. Due to the computational complexity and for the improvement of the model efficiency, the dimensionality reduction technique, Principal Component Analysis (PCA) was then applied. This resulted in an optimum representation of 100 principal components, retaining about 96% of the variance. Then, the performances of various machine learning classifiers such as k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Decision Tree (DT), Neural Network (NN), Naïve Bayes (NB), Logistic Regression (LR) and Enhanced Fuzzy Min–Max Neural Network (EFMM) were tested. The accuracy, precision, recall, F1-score, area under the receiver operating characteristic curve (AUC), and confusion matrices were used to evaluate the performance. Stratified 5-fold cross validation was used to ensure the strength of our results. Results: The findings show that PCA has a significant improvement in classification accuracy for most of the models. In particular, the EFMM classifier outperformed the other classifiers, with an accuracy of 97.19% on the independent test set. After PCA, the AUC values for classes such as Mild Demented, Moderate Demented, Non-Demented and Very Mild Demented were obtained as 97.12%, 99.97%, 93.79% and 95.26% respectively. We further validated our proposed framework using stratified 5-fold cross validation which further corroborated the robustness of our proposed framework. The EFMM achieved a mean accuracy of 98.38% ± 0.36 and a mean macro-F1 score of 98.48% ± 0.43. Friedman statistical testing demonstrated that there were significant differences between the performance of the classifiers evaluated (p < 0.001), which further validated the performance of the EFMM. Conclusions: To sum up, the proposed SqueezeNet–PCA–EFMM is an effective and efficient method for Alzheimer’s disease stage classification under MRI images. The combination of SqueezeNet, PCA, and EFMM—led not only to high classification performance, but also to good cross validation results. Furthermore, this property of incremental learning is the intrinsic one of the EFMM and renders this framework interesting for its incorporation in the next-generation intelligent clinical decision supports in particular, as medical care evolves. Full article

Accurate water saturation estimation in deep shale reservoirs is complicated by clay-related additional conductivity and coupled pore, organic-matter, and structural effects. This study develops a feature-level physics-guided machine-learning framework, termed PhysML-Hybrid. Five mechanism-derived descriptor groups representing clay–water interfacial behavior, low-resistivity correction, pore connectivity, organic-pore development, and structural stress were integrated with conventional reservoir variables in a validation-weighted ensemble of random forest, XGBoost, and Bayesian neural network models. The framework was evaluated using 153 depth-matched samples from five wells in the Dingshan area of the Sichuan Basin. The data were divided into 107 training, 16 validation, and 30 independent test samples, and target-stratified five-fold cross-validation was conducted exclusively within the training set. Mean cross-validation R², MAE, and RMSE were 0.907 ± 0.009, 1.69% ± 0.10%, and 2.25% ± 0.14%, respectively. On the independent test set, the corresponding values were 0.902, 1.77%, and 2.34%. PhysML-Hybrid outperformed Archie, SVM, ML-only, and Phy-XGB. SHAP and statistical analyses identified clay content, the curvature–clay interaction, TOC, pore connectivity, and structural descriptors as influential variables; candidate transitions were interpreted as dataset-specific rather than universal thresholds or causal relationships. Three blind-well cases provided supplementary evidence of cross-well applicability, although larger independent multi-basin datasets are required to assess transferability. Full article

Nuclei instance segmentation in hematoxylin and eosin (H&E)-stained images plays an important role in automated histological image analysis, with various applications in downstream tasks. While several machine learning and deep learning approaches have been proposed for nuclei instance segmentation, most research in this field focuses on developing new segmentation algorithms and benchmarking them on a limited number of arbitrarily selected public datasets. In this work, rather than focusing on model development, we focused on the datasets used for this task. Based on an extensive literature review, we identified manually annotated, publicly available datasets of H&E-stained images for nuclei instance segmentation and standardized them into a unified input and annotation format. Using two state-of-the-art segmentation models, one based on convolutional neural networks (CNNs) and one based on a hybrid CNN and vision transformer architecture, we systematically evaluated and ranked these datasets based on their nuclei instance segmentation performance. Furthermore, we proposed a unified test set (NucFuse-test) for fair cross-dataset evaluation and a unified training set (NucFuse-train) for improved segmentation performance by merging images from multiple datasets. To the best of our knowledge, this is the first study to systematically benchmark and rank publicly available datasets for nuclei instance segmentation. By evaluating and ranking the datasets, performing comprehensive analyses, generating fused datasets, conducting external validation, and making our implementation publicly available, we provided a new protocol for training, testing, and evaluating nuclei instance segmentation models on H&E-stained histological images. Full article

The integration of artificial intelligence (AI) with coagulation–flocculation (C-F) processes represents a significant advancement for optimizing wastewater treatment. However, a comprehensive analysis of the research landscape, trends and collaborative networks in this interdisciplinary field remains lacking. This study addresses this gap by conducting a bibliometric analysis of 251 Scopus-indexed publications (2000–2025) using VOSviewer and Bibliometrix. The objective was to map the intellectual structure, quantify growth trends and identify key research themes and contributors. Results indicate a surge in publications post-2015, with environmental science and engineering as dominant subject areas. China, Iran and Nigeria are leading contributors though geographical concentration suggests a need for broader collaboration. Keyword analysis reveals a thematic evolution from basic artificial neural networks towards advanced machine learning and deep learning, primarily focused on optimizing coagulant dosage and predictive control. Despite promising advancements, challenges related to data dependency, model interpretability and infrastructure integration persist. Future prospects hinge on developing explainable and hybrid AI models, leveraging the Internet of Things for real-time adaptation and fostering interdisciplinary research to bridge the gap between data science and process engineering. This analysis provides a foundational overview to guide future research towards more robust, transparent and widely applicable AI-driven C-F systems in sustainable wastewater management. Full article

Background: The accurate enumeration and identification of probiotic strains are essential for product quality. The plate count (PC) gold standard enumerates viable, culturable cells but does not by itself resolve individual strains within multi-strain consortia, and molecular methods (qPCR, ddPCR) are costly and face recognised challenges in quantifying relative strain abundance. Fourier transform infrared (FTIR) spectroscopy is a promising phenotypic alternative. Methods: We developed an FTIR-based artificial neural network classifier to identify and quantify a four-strain probiotic blend comprising Lactobacillus acidophilus LA02, Lacticaseibacillus rhamnosus LR04, Limosilactobacillus fermentum LF08, and Bifidobacterium animalis subsp. lactis BS01 compared against selective plate counting and species-specific PCR. Results: The classifier correctly identified all 36 test colonies (100%; 95% Clopper–Pearson CI: 90.3–100%); descriptive cluster analysis indicated spectral distinctiveness (silhouette = 0.908; Davies–Bouldin = 0.127; cophenetic correlation = 0.955). Enumeration agreement with selective plate counting was assessed descriptively (Pearson r = 0.78, 95% CI [−0.72, 1.00], n = 4 strain means; mean difference −0.028 log₁₀ CFU/mL; all strain-level differences < 0.1 log₁₀ CFU/mL). PCR confirmed all FTIR classifications. Conclusions: This proof-of-concept study demonstrates the feasibility of coupling cultivation with spectroscopic identification in a hybrid PC + FTIR workflow for multi-strain probiotic quality control. Because the findings derive from a single blend preparation analysed in technical replicates, they characterise this dataset and require confirmation on independently prepared batches before the approach can be regarded as a validated method. Full article

Bio-jet fuels produced from biomass-derived feedstocks represent a promising pathway for reducing the carbon intensity of aviation energy systems. However, designing supply chain networks for bio-jet fuel production requires solving large-scale mixed-integer linear programming (MILP) models that integrate facility location, feedstock allocation, material flows, and routing decisions. These models can become computationally expensive, particularly when evaluating multiple network configurations or large candidate sets of production and processing facilities. This study develops a hybrid machine learning and optimization framework to improve the computational efficiency of bio-jet fuel supply chain network design while preserving high-quality decision outcomes. The proposed iterative procedure uses supervised learning to approximate the relationship between facility location decisions and total supply chain cost. First, an initial set of supply chain configurations is generated by solving the optimization model while using randomly selected facility locations. These solutions are then used to train predictive models, including ridge regression, feed-forward neural networks, and ensemble neural networks, with facility-location configurations as inputs and total supply chain cost as the output. The trained learner is subsequently used to identify promising facility-location candidates through Thompson sampling and small-scale linear programming. These candidate solutions are evaluated by the original mixed-integer model, and the resulting observations are fed back into the learning process until convergence. Numerical experiments show that the proposed hybrid approach obtains near-optimal bio-jet fuel supply chain designs while substantially reducing computational time. For the linear case, the method achieves solutions within 0.23–0.29% of the objective function value while reducing computational time by 70.95–81.95%. For nonlinear learning models, the optimality gap decreases further to 0.13–0.15%, with computational time reductions of 45.37–60.36%. For the Texas case study and the modeling assumptions evaluated, the findings demonstrate that machine-learning-assisted optimization can reduce computational effort while preserving high-quality supply chain solutions. The extent of these benefits may vary with network size, candidate-facility structure, facility-capacity assumptions, demand characteristics, and the amount of information available to train the learning models. Full article