Search Results (8,276)

Efficiently identifying suitable solvents for active pharmaceutical ingredients (APIs) is critical in drug formulation, yet the vast number of possible solvent-solute combinations presents a significant experimental challenge. This study addresses this by developing a robust machine learning (ML) model for accurately predicting the solubility of three phenolic acids (syringic, p-coumaric, and caffeic) in various deep eutectic solvents (DESs), integrating both experimental and computational investigations. Measured solubility data showed that the choline chloride combined with triethylene glycol in a 1:2 molar ratio was the most efficient system for the dissolution of the studied APIs. Different ML models, utilizing nu-Support Vector Regression (nuSVR) as the core regressor and based on descriptor sets derived from COSMO-RS (Conductor-like Screening Model for Real Solvents) computations, were systematically evaluated. A novel methodology termed DOO-IT (Dual-Objective Optimization with ITerative feature pruning) was employed to address the common challenges of model development with limited, high-value datasets. The final optimal 10-descriptor nuSVR model, selected from an exhaustive, multi-run search, demonstrated outstanding predictive power, offering a highly reliable computational tool for guiding experimental screening, significantly accelerating the exploration of DES-based formulations. This research also provides a strong foundation for future machine learning-guided discovery of chemicals, offering an effective and transferable framework for developing QSPR models for various chemical systems. Full article

Background/Objectives: Accurate identification of deciduous and permanent teeth on panoramic radiographs (PRs) during mixed dentition is fundamental for early detection of eruption disturbances, yet relies heavily on clinician experience due to developmental variability. This study aimed to develop a deep learning model for automated tooth detection and segmentation in pediatric PRs during mixed dentition. Methods: A retrospective dataset of 250 panoramic radiographs from patients aged 6–13 years was analyzed. A customized YOLOv11-based model was developed using a novel hybrid pre-annotation strategy leveraging transfer learning from 650 publicly available adult radiographs, followed by expert manual refinement. Performance evaluation utilized mean average precision (mAP), F1-score, precision, and recall metrics. Results: The model demonstrated robust performance with mAP_0.5 = 0.963 [95%CI: 0.944–0.983] and macro-averaged F1-score = 0.953 [95%CI: 0.922–0.965] for detection. Segmentation achieved mAP_0.5 = 0.890 [95%CI: 0.857–0.923]. Stratified analysis revealed excellent performance for permanent teeth (F1 = 0.977) and clinically acceptable accuracy for deciduous teeth (F1 = 0.884). Conclusions: The automated system achieved near-expert accuracy in detecting and segmenting teeth during mixed dentition using an innovative transfer learning approach. This framework establishes reliable infrastructure for AI-assisted diagnostic applications targeting eruption or developmental anomalies, potentially facilitating earlier detection while reducing clinician-dependent variability in mixed dentition evaluation. Full article

Traditional empirical correlations for predicting saturated flow boiling heat transfer coefficients (HTC) often struggle with accuracy and generalizability, particularly across different refrigerants, heat exchanger geometries, and operating conditions. To address these limitations, this study investigates the application of machine learning for more robust HTC prediction. A comprehensive dataset was compiled, consisting of 22,608 data points from over 140 published studies, covering 18 pure refrigerants under diverse experimental setups. The primary goal was to evaluate the performance of different machine learning approaches—Wide Neural Network (WNN), Linear Regression (LR), and Support Vector Machine (SVM)—in predicting HTCs across varying tube types and heat exchanger configurations. The results indicate that the WNN model achieved the highest predictive accuracy, with a Root Mean Square Error (RMSE) of 1.97 and a coefficient of determination (R²) of 0.91, corresponding to less than 5% prediction error for all refrigerants. These outcomes confirm that machine learning models can effectively capture the complex thermofluid interactions involved in boiling heat transfer. This work demonstrates that data-driven methods provide a reliable and generalizable alternative to empirical correlations. Full article

Water quality is an essential part of maintaining a healthy environment for fish farming. The quality of the water is related to a few of the chemical and biological characteristics of water. The conventional evaluation methods of the water quality are often time-consuming and may overlook complex interdependencies among multiple indicators. This study has proposed a robust machine learning framework for aquaculture water quality classification by integrating the Honey Badger Algorithm (HBA) with the XGBoost classifier. The framework enhances classification accuracy and incorporates explainability through SHAP and DiCE, thereby providing both predictive performance and transparency for practical water quality management. For reliability, the dataset has been randomly shuffled, and a custom 5-fold cross-validation strategy has been applied. Later, through the metaheuristic-based HBA, feature selections and hyperparameter tuning have been performed to improve and increase the prediction accuracy. The highest accuracy of 98.45% has been achieved by a particular fold, whereas the average accuracy is 98.05% across all folds, indicating the model’s stability. SHAP analysis reveals Ammonia, Nitrite, DO, Turbidity, BOD, Temperature, pH, and CO₂ as the topmost water quality indicators. Finally, the DiCE analysis has analyzed that Temperature, Turbidity, DO, BOD, CO₂, pH, Ammonia, and Nitrite are more influential parameters of water quality. Full article

Temporal network diffusion models play a crucial role in healthcare, information technology, and machine learning, enabling the analysis of dynamic event-based processes such as disease spread, information propagation, and behavioral diffusion. This study addresses the challenge of reconstructing temporal network diffusion events in real time under conditions of missing and evolving data. A novel non-parametric reconstruction method by simple weights differentiationis proposed to enhance source detection robustness with provable improved error bounds. The approach introduces adaptive cost adjustments, dynamically reducing high-risk source penalties and enabling bounded detours to mitigate errors introduced by missing edges. Theoretical analysis establishes enhanced upper bounds on false positives caused by detouring, while a stepwise evaluation of dynamic costs minimizes redundant solutions, resulting in robust Steiner tree reconstructions. Empirical validation on three real-world datasets demonstrates a 5% improvement in Matthews correlation coefficient (MCC), a twofold reduction in redundant sources, and a 50% decrease in source variance. These results confirm the effectiveness of the proposed method in accurately reconstructing temporal network diffusion while improving stability and reliability in both offline and online settings. Full article

Prior studies on indoor LiDAR point-cloud semantic segmentation consistently report that sampling density strongly affects segmentation accuracy as well as runtime and memory, establishing an accuracy–efficiency trade-off. Nevertheless, in practice, the density is often chosen heuristically and reported under heterogeneous protocols, which limits quantitative guidance. We present a unified evaluation framework that treats density as the sole independent variable. To control architectural variability, three representative backbones—PointNet, PointNet++, and DGCNN—are each augmented with an identical Point Transformer module, yielding PointNet-Trans, PointNet++-Trans, and DGCNN-Trans trained and tested under one standardized protocol. The framework couples isotropic voxel-guided uniform down-sampling with a decision rule integrating three signals: (i) accuracy sufficiency, (ii) the onset of diminishing efficiency, and (iii) the knee of the accuracy–density curve. Experiments on scan-derived indoor point clouds (with BIM-derived counterparts for contrast) quantify the accuracy–runtime trade-off and identify an engineering-feasible operating band of 1600–2900 points/m², with a robust setting near 2400 points/m². Planar components saturate at moderate densities, whereas beams are more sensitive to down-sampling. By isolating density effects and enforcing one protocol, the study provides reproducible, model-agnostic guidance for scan planning and compute budgeting in indoor mapping and Scan-to-BIM workflows. Full article

Food legumes play a pivotal role in China’s food security, nutritional health, and green development strategies due to their unique advantages. This paper presents an empirical study on the economic evaluation of scientific research on pest and disease control for food legumes. Using panel data from 31 Chinese provinces from 2008 to 2023, we employ a Double Machine Learning (DML) approach to identify the impact of investment in plant protection research on food legume outputs. The results indicate a steady increase in China’s investment in this field, with an average annual growth rate of 5.19% from 2008 to 2023, and the total investment in 2023 was 2.14 times that of 2008. Investment in plant protection research effectively mitigates output losses and leads to significant production increases. Specifically, a 1% increase in research investment corresponds to a 0.2% increase in food legume output. This effect remains robust across various algorithms, time windows, and control variable settings. Based on these findings, we recommend: (1) increasing financial support and talent acquisition for research on food legume pests and diseases to enhance the stability and sustainability of research investment; (2) strengthening cooperation mechanisms between research institutions and enterprises to leverage their respective strengths and promote the commercialization of research outcomes and regional variety extension; (3) establishing a diversified research investment system that explores a co-construction model guided by the government, involving enterprises, and utilizing public–private partnerships to reconcile the conflict between long research cycles and market demands; (4) fostering a dual-track linkage between regional technological innovation and enterprise product commercialization to improve the efficiency of technology transfer and application; and (5) strengthening R&D in cutting-edge fields like Artificial Intelligence to improve the efficiency and precision of pest and disease control. Full article

This paper presents a dual-objective Model Predictive Control (MPC) framework for fixed-wing unmanned aerial vehicles (UAVs). The framework was designed with two goals in mind: improving longitudinal motion control and optimizing the flight trajectory when connectivity and no-fly zone constraints are present. A multi-input–multi-output model derived from NASA’s Generic Transport Model (T-2) was used and linearized for controller design. We compared the MPC controller with a Linear Quadratic Regulator (LQR) in MATLAB simulations. The results showed that MPC reached the reference values faster, with less overshoot and phase error, particularly under sinusoidal reference inputs. These differences became even more evident when the UAV had to fly in windy conditions. Trajectory optimization was carried out using the CasADi framework, which allowed us to evaluate paths that balance two competing requirements: reaching the target quickly and maintaining cellular connectivity. We observed that changing the weights of the cost function had a strong influence on the trade-off between direct flight and reliable communication, especially when multiple base stations and no-fly zones were included. Although the study was limited to simulations at constant altitude, the results suggest that MPC can serve as a practical tool for UAV missions that demand both accurate flight control and robust connectivity. Future work will extend the framework to more complete models and experimental validation. Full article

Industrial safety in high-risk sectors such as mining, construction, oil and gas, petrochemicals, and offshore fishing remains a strategic global challenge due to the high incidence of occupational accidents and their human, financial, and legal consequences. Despite international standards and advancements in safety strategies, significant barriers persist in the effective implementation of a Zero Accident culture. This scoping review, conducted under PRISMA-ScR guidelines, analyzed 11 studies selected from 232 records, focusing on documented practices in both multinational corporations from developed economies and local companies in emerging markets. The methodological synthesis validated theoretical models, practical interventions, and regulatory frameworks across diverse industrial settings. The findings led to the construction of a five-pillar model that provides the structural foundation for a comprehensive safety strategy: (1) strategic safety planning, defining long-term vision, mission, and objectives with systematic risk analysis; (2) executive leadership and commitment, expressed through decision-making, resource allocation, and on-site engagement; (3) people and competencies, emphasizing continuous training, communities of practice, and the development of safe behaviors; (4) process risk management, using validated protocols, structured methodologies, and early warning systems; and (5) performance measurement and auditing, combining reactive and proactive indicators within continuous improvement cycles. The results demonstrate that only a holistic approach, one that aligns strategy, culture, and performance, can sustain a robust safety culture. While notable reductions in incident rates were observed when these pillars were applied, the current literature is dominated by theoretical contributions and model replication from developed countries, with limited empirical evaluation in emerging contexts. This study provides a comparative, practice-oriented framework to guide the implementation and refinement of safety systems in high-risk organizations. This review was registered in Open Science Framework (OSF): 10.17605/OSF.IO/XFDPR. Full article

Designing Low Earth Orbit (LEO) constellations for the continuous, collaborative observation of space objects in MEO/GEO is a complex optimization task, frequently limited by prohibitive computational costs. This study introduces an efficient surrogate-based framework to overcome this challenge. Our approach integrates Optimized Latin Hypercube Sampling (OLHS) with a Radial Basis Function (RBF) model to minimize the required number of satellites. In a comprehensive case study targeting 18 diverse space objects—including communication satellites in GEO (e.g., EUTELSAT, ANIK) and navigation satellites in MEO/IGSO from GPS, Galileo, and BeiDou constellations—the method proved highly effective and scalable. It successfully designed a 208-satellite Walker constellation that provides 100% continuous coverage over a 36-h period. Furthermore, the design ensures that each target is simultaneously observed by at least three satellites at all times. A key finding is the method’s remarkable efficiency and scalability: the optimal solution for this larger problem was found using only 46 high-fidelity function evaluations, maintaining a computational time that was 5–8 times faster than traditional global optimization algorithms. This research demonstrates that surrogate-assisted optimization can drastically lower the computational barrier in constellation design, offering a powerful tool for building cost-effective and robust Space Situational Awareness (SSA) systems. Full article

The development of anticancer peptides (ACPs) has emerged as a promising strategy in targeted cancer therapy due to their high specificity and therapeutic potential. Cathelicidin-derived antimicrobial peptides represent a particularly attractive class of ACPs, yet systematic evaluation of their anticancer activity remains limited. In this study, we conducted virtual screening of eight cathelicidin-derived peptides (AL-38, LL-37, RK-31, KS-30, KR-20, FK-16, FK-13, and KR-12) to assess their potential against colon cancer. Among these, LL-37 and FK-16 were identified as the most promising candidates, with LL-37 exhibiting the strongest inhibitory effects on both non-metastatic (HT-29) and metastatic (SW-620) colon cancer cell lines in vitro. To overcome challenges associated with peptide stability and delivery, we employed the probiotic lactic acid bacterium Limosilactobacillus fermentum KUB-D18 as both a biosynthetic platform and delivery vehicle. A genome-scale metabolic model (GEM), iTM505, was reconstructed to predict the strain’s biosynthetic capacity for ACP production. Model simulations identified trehalose, sucrose, maltose, and cellobiose as optimal carbon sources supporting both high peptide yield and biomass accumulation, which was subsequently confirmed experimentally. Notably, L. fermentum expressing LL-37 achieved a growth rate of 2.16 gDW/L, closely matching the model prediction of 1.93 gDW/L (accuracy 89.69%), while the measured LL-37 concentration (26.96 ± 0.08 µM) aligned with predictions at 90.65% accuracy. The strong concordance between in silico predictions and experimental outcomes underscore the utility of GEM-guided metabolic engineering for optimizing peptide biosynthesis. This integrative approach—combining virtual screening, genome-scale modeling, and experimental validation—provides a robust framework for accelerating ACP discovery. Moreover, our findings highlight the potential of probiotic-based systems as effective delivery platforms for anticancer peptides, offering new avenues for the rational design and production of peptide therapeutics. Full article

Afforestation is a critical nature-based strategy for enhancing ecological resilience and supporting cleaner land-use systems. This study presents a spatially explicit modeling framework to evaluate the long-term impacts of potential afforestation amendments on two key ecosystem services—soil conservation and carbon sequestration—across China’s major shelterbelt program areas under the SSP245 scenario (2020–2070). Using a zonal approach, we integrated Random Forest models, Bayesian belief networks, and Geodetector analysis to identify region-specific afforestation suitability and quantify ecological service gains across eight national shelterbelt program zones. The results reveal pronounced spatial heterogeneity in ecosystem service improvements. (1) High-quality potential afforestation lands, totaling approximately 2.33 × 10⁵ km², are primarily concentrated near the Hu Line (a geographical boundary that divides China into two distinct climatic regions), with the shelterbelt program for upper and middle reaches of Yangtze River accounting for 45.94%. (2) Based on the amended annual afforestation target of 0.47 × 10⁵ km², the adjusted land use projections indicate a significant increase in forest cover. By 2070, the afforestation program for Taihang Mountain exhibits the most significant improvements, with a 47.56% increase in soil conservation and a 10.15% increase in carbon sequestration. (3) Optimization areas differ across zones, with the Taihang mountain area (99.2%) and Pearl river area (70.1%) achieving the highest improvements in soil and carbon services, respectively. These findings provide robust scientific support for data-driven, region-specific afforestation planning under future land-use change scenarios. Full article

Background/Objectives: Chronic kidney disease (CKD) is a progressive condition that affects the body’s ability to remove waste and regulate fluid and electrolytes. Early detection is crucial for delaying disease progression and initiating timely interventions. Machine learning (ML) techniques have emerged as powerful tools for automating disease diagnosis and prognosis. This study aims to evaluate the predictive performance of individual and ensemble ML algorithms for the early classification of CKD. Methods: A clinically annotated dataset was utilized to categorize patients into CKD and non-CKD groups. The models investigated included Logistic Regression, Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Ridge Classifier, Naïve Bayes, K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Ensemble learning strategies. A systematic preprocessing pipeline was implemented, and model performance was assessed using accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve (AUC). Results: The empirical findings reveal that ML-based classifiers achieved high predictive accuracy in CKD detection. Ensemble learning methods outperformed individual models in terms of robustness and generalization, indicating their potential in clinical decision-making contexts. Conclusions: The study demonstrates the efficacy of ML-based frameworks for early CKD prediction, offering a scalable, interpretable, and accurate clinical decision support approach. The proposed methodology supports timely diagnosis and can assist healthcare professionals in improving patient outcomes. Full article

To enable effective change impact management in business applications, robust requirements traceability is essential. However, manual approaches are inefficient and prone to errors. While the prior Model-Driven Engineering (MDE)-based research, including the author’s theoretical models, established the principles of traceability, these approaches lacked decisive quantitative validation using metrics such as precision and recall, thereby limiting their real-world applicability. This paper addresses these limitations by introducing the CRSensor framework, which integrates the real-time automated trace link generation and dynamic refinement of the developer model. This approach enhances the reliability and completeness of organizational impact analysis, resolving key weaknesses of conventional link recovery methods. Notably, CRSensor maintains structural consistency throughout the lifecycle, overcoming reliability limitations often found in traditional information retrieval (IR)/machine learning (ML)-based traceability solutions. Empirical evaluation demonstrates that CRSensor achieves an average trace link setting performance with a precision of 0.95, a recall of 0.98, and an auto-generation rate of 80%. These results validate both the industrial applicability and the quantitative rigor of the proposed framework, paving the way for broader practical adoption. Full article

Background: Accurately detecting the onset of saturated boiling in herbal medicine extraction processes is critical for improving production efficiency and reducing energy consumption. However, the traditional monitoring methods based on temperature suffer from time delays. To address the challenge, acoustic emission (AE) signals were used in this study owing to its sensitivity to bubble behavior. Methods: An AE signal acquisition system was constructed for herbal extraction monitoring. Characteristics of AE signals at different boiling stages were analyzed in pure water systems with and without herbs. The performance of AE-based and temperature-based recognition of boiling stages was compared. To enhance applicability in different herb extraction systems, multivariate statistical analysis was adopted to compress spectral–frequency information into Hotelling’s T² and SPE statistics. For real-time monitoring, a self-adaptive decision-making boiling judgment method (BoilStart) was proposed. To evaluate the robustness, the performance of BoilStart under different conditions was investigated, including extraction system mass and heating medium temperature. Furthermore, BoilStart was applied to a lab-scale extraction process of Dabuyin Wan, which is a practical formulation, to assess its performance in energy conservation and efficiency improvement. Results: AE signal in the 75–100 kHz frequency band could reflect the boiling states of herbal medicine extraction. It was more sensitive to the onset of saturated boiling than the temperature signal. Compared with SPE, Hotelling’s T² was identified as the optimal indicator with higher accuracy. BoilStart could adaptively monitor saturated boiling across diverse herbal systems. The absolute error of BoilStart’s boiling determination ranged from 1.5 min to 2.0 min. The increasing-temperature time was reduced by about 22–36%. For the extraction process of Dabuyin Wan, after adopting BoilStart, the increasing-temperature time was reduced by about 29%, and the corresponding energy consumption was lowered by about 26%. Conclusions: The first AE-based method for precise boiling state detection in herbal extraction was established. BoilStart’s model-free adaptability met industrial demands for multi-herb compatibility. This offered a practical solution to shorten ineffective heating phases and reduce energy consumption. Full article