Search

As a promising strategy to mitigate carbon emissions, hydrogen enrichment of conventional fuels is gaining increasing interest, but can lead to increased propensity to damaging thermoacoustic instability. This paper demonstrates the ability of machine learning algorithms to reliably predict a key flame behavior relevant to thermoacoustic stability prediction – the flame's nonlinear response to upstream acoustic forcing. Using computational simulations of a laminar premixed flame under difference hydrogen enrichment and equivalence ratio conditions, training data for machine learning are generated. The machine learning models are then used to predict the flame response under new test conditions, including extrapolating into new hydrogen enrichment regimes. The effect of algorithm choice, data structure, data size, and extrapolation distance on the performance of different machine learning models is investigated with particular attention to performance with sparse training data. It is found that a type of neural network known as the multi-layer perceptron (MLP) model outperforms the Gaussian process and random forest models in both interpolation and extrapolation tasks. A physics-informed preprocessing strategy, involving extracting the time delay of the flame response (which can be deduced from geometry and flow parameters) prior to applying machine learning, is found to remarkably improve the accuracy of machine learning models, especially in extrapolation tasks. When the extrapolation into new hydrogen content levels is strong enough, the MLP model no longer performs well. However, for small and intermediate extrapolation, the MLP model shows a commendable level of prediction, even with small data sets, fulfilling needs relevant to thermoacoustic stability prediction. Extending the MLP model to experimental data from turbulent flames, the gain fall-off frequency and phase lag are accurately predicted but not the secondary periodic fluctuations in gain. • AI-assisted prediction of flame response in new hydrogen enrichment regimes. • A physics-informed strategy extracting the time delay improves model performance. • The effects of algorithm choice, data structure and data size are studied. • MLP model shows good predictions in small extrapolation tasks with sparse data. • Extending to turbulent flames, the MLP model captures fall-off frequency in gain. [ABSTRACT FROM AUTHOR]

Background: The congenital ventricular outflow tract malformations (CVOTMs) is a major congenital heart diseases (CHDs) subtype, and its pathogenesis is complex and unclear. Lipid metabolic plays a crucial role in embryonic cardiovascular development. However, due to the limited types of detectable metabolites in previous studies, findings on lipid metabolic and CHDs are still inconsistent, and the possible mechanism of CHDs remains unclear. Methods: The nest case-control study obtained subjects from the multicenter China Teratology Birth Cohort (CTBC), and maternal serum from the pregnant women enrolled during the first trimester was utilized. The subjects were divided into a discovery set and a validation set. The metabolomics of CVOTMs and normal fetuses were analyzed by targeted lipid metabolomics. Differential comparison, random forest and lasso regression were used to screen metabolic biomarkers. Results: The lipid metabolites were distributed differentially between the cases and controls. Setting the selection criteria of P value < 0.05, and fold change (FC) > 1.2 or < 0.833, we screened 70 differential metabolites. Within the prediction model by random forest and lasso regression, DG (14:0_18:0), DG (20:0_18:0), Cer (d18:2/20:0), Cer (d18:1/20:0) and LPC (0:0/18:1) showed good prediction effects in discovery and validation sets. Differential metabolites were mainly concentrated in glycerolipid and glycerophospholipids metabolism, insulin resistance and lipid & atherosclerosis pathways, which may be related to the occurrence and development of CVOTMs. Conclusion: Findings in this study provide a new metabolite data source for the research on CHDs. The differential metabolites and involved metabolic pathways may suggest new ideas for further mechanistic exploration of CHDs, and the selected biomarkers may provide some new clues for detection of COVTMs. [ABSTRACT FROM AUTHOR]

In this study, we evaluated the performance of various machine-learning models on multiple datasets labeled GR1, GR2, GR3, GR4, and GR5. We assessed the models using a range of evaluation metrics, including AUC, CA, F1, precision, recall, MCC, specificity, and log loss. The models examined were logistic regression, decision tree, kNN, random forest, gradient boosting, neural network, AdaBoost, and stochastic gradient descent. The results indicate that all models consistently demonstrated outstanding performance across all datasets, with most achieving perfect scores in all metrics. The models exhibited high accuracy and effectiveness in accurately classifying instances. Although random forests displayed slightly lower scores in some metrics, theyi still maintained an overall high level of accuracy. The findings highlight the models' ability to effectively learn the underlying patterns within the data and make accurate predictions. The low log loss values further confirmed the models' precise estimation of probabilities. Consequently, these models possess strong potential for practical applications in various domains, offering reliable and robust classification capabilities. [ABSTRACT FROM AUTHOR]

Cancer ranks among the most lethal illnesses worldwide, and predicting its onset can be a crucial factor in enhancing people's quality of life by taking preventive measures to improve treatment and survival. This study conducted comparative research to determine the machine learning model with the highest accuracy for tumor type classification, distinguishing between malignant (cancer) and benign tumors. The models evaluated include decision tree (DT), naive bayes (NB), extra trees classifier (ETM), random forest (RF), K-means clustering (K-means), logistic regression (LR), adaptive boosting (AdaBoost), gradient boosting (GB), light gradient boosting machine (LightGBM), and extreme gradient boosting (XGBoost) to identify the one with the best accuracy. The models were trained using a dataset of 569 records and a total of 32 variables, containing patient information and tumor characteristics. The study was structured into sections, such as related studies, descriptions of the models, case study development, results, discussion, and conclusions. The models' performance was evaluated based on metrics of precision, sensitivity, accuracy, and F1 score. Following the training, the results positioned the XGBoost model as having the best performance, achieving 98% precision, accuracy, sensitivity, and F1 score. [ABSTRACT FROM AUTHOR]

Land Use and Land Cover (LULC) maps are important tools for environmental planning and social-ecological modeling, as they provide critical information for evaluating risks, managing natural resources, and facilitating effective decision-making. This study aimed to generate a very high spatial resolution (0.5 m) and detailed (21 classes) LULC map for the greater Mariño watershed (Peru) in 2019, using the MORINGA processing chain. This new method for LULC mapping consisted in a supervised object-based LULC classification, using the random forest algorithm along with multi-sensor satellite imagery from which spectral and textural predictors were derived (a very high spatial resolution Pléiades image and a time serie of high spatial resolution Sentinel-2 images). The random forest classifier showed a very good performance and the LULC map was further improved through additional post-treatment steps that included cross-checking with external GIS data sources and manual correction using photointerpretation, resulting in a more accurate and reliable map. The final LULC provides new information for environmental management and monitoring in the greater Mariño watershed. With this study we contribute to the efforts to develop standardized and replicable methodologies for high-resolution and high-accuracy LULC mapping, which is crucial for informed decision-making and conservation strategies. [ABSTRACT FROM AUTHOR]

Snow avalanches are the primary mountain hazard for mechanized skiing operations. Helicopter and snowcat ski guides are tasked with finding safe terrain to provide guests with enjoyable skiing in a fast-paced and highly dynamic and complex decision environment. Based on years of experience, ski guides have established systematic decision-making practices that streamline the process and limit the potential negative influences of time pressure and emotional investment. While this expertise is shared within guiding teams through mentorship, the current lack of a quantitative description of the process prevents the development of decision aids that could strengthen the process. To address this knowledge gap, we collaborated with guides at Canadian Mountain Holidays (CMH) Galena Lodge to catalogue and analyze their decision-making process for the daily run list, where they code runs as green (open for guiding), red (closed), or black (not considered) before heading into the field. To capture the real-world decision-making process, we first built the structure of the decision-making process with input from guides, and then used a wide range of available relevant data indicative of run characteristics, current conditions, and prior run list decisions to create the features of the models. We employed three different modelling approaches to capture the run list decision-making process: Bayesian Network, Random Forest, and Extreme Gradient Boosting. The overall accuracies of the models are 84.6 %, 91.9 %, and 93.3 % respectively, compared to a testing dataset of roughly 20,000 observed run codes. The insights of our analysis provide a baseline for the development of effective decision support tools for backcountry avalanche risk management that can offer independent perspectives on operational terrain choices based on historic patterns or as a training tool for newer guides. [ABSTRACT FROM AUTHOR]

Summary The increasing demand for massaging networks that are stable and quick needs reevaluations of standard optical networking administration strategies. To improve the efficacy of optical networks by integrating machine learning (ML) approach for the best resource scheduling, this research presents an innovative dynamic block widow optimized random forest (DBWO‐RF) strategy. To implement the DBWO‐driven resource allocation method in accordance with the categorization and clustering findings, the RF method is incorporated with the software defined optical to achieve channel quality assessment after successfully clustering employs the RF approach to achieve channel quality assessment after successfully clustering traffic patterns using the fuzzy C‐means (FCM) algorithm. To lessen the likelihood of blocking, the fragmentation‐function‐fit (FFF) algorithm was provided and the findings indicate that this approach possesses a reduced blocking risk. Using multiple approaches to modulation for various channel quality, the suggested resource allocation system leverages the DBWO approach to distribute the necessary resources based on various “traffic flow (TF)” clustering findings. The examination's outcomes demonstrate that, compared to other techniques under various given load levels, the present study has a reduced blocking risk, a sufficient complexity degree and greater effectiveness in the utilization of spectrum resources. [ABSTRACT FROM AUTHOR]

Purpose: To identify the key risk factors for venous thromboembolism (VTE) in urological inpatients based on the Caprini scale using an interpretable machine learning method. Methods: VTE risk data of urological inpatients were obtained based on the Caprini scale in the case hospital. Based on the data, the Boruta method was used to further select the key variables from the 37 variables in the Caprini scale. Furthermore, decision rules corresponding to each risk level were generated using the rough set (RS) method. Finally, random forest (RF), support vector machine (SVM), and backpropagation artificial neural network (BPANN) were used to verify the data accuracy and were compared with the RS method. Results: Following the screening, the key risk factors for VTE in urology were "(C1) Age," "(C2) Minor Surgery planned," "(C3) Obesity (BMI > 25)," "(C8) Varicose veins," "(C9) Sepsis (< 1 month)," (C10) "Serious lung disease incl. pneumonia (< 1month) " (C11) COPD," "(C16) Other risk," "(C18) Major surgery (> 45 min)," "(C19) Laparoscopic surgery (> 45 min)," "(C20) Patient confined to bed (> 72 h)," "(C18) Malignancy (present or previous)," "(C23) Central venous access," "(C31) History of DVT/PE," "(C32) Other congenital or acquired thrombophilia," and "(C34) Stroke (< 1 month." According to the decision rules of different risk levels obtained using the RS method, "(C1) Age," "(C18) Major surgery (> 45 minutes)," and "(C21) Malignancy (present or previous)" were the main factors influencing mid- and high-risk levels, and some suggestions on VTE prevention were indicated based on these three factors. The average accuracies of the RS, RF, SVM, and BPANN models were 79.5%, 87.9%, 92.6%, and 97.2%, respectively. In addition, BPANN had the highest accuracy, recall, F1-score, and precision. Conclusions: The RS model achieved poorer accuracy than the other three common machine learning models. However, the RS model provides strong interpretability and allows for the identification of high-risk factors and decision rules influencing high-risk assessments of VTE in urology. This transparency is very important for clinicians in the risk assessment process. [ABSTRACT FROM AUTHOR]

Air quality is a fundamental component of a healthy environment for human beings. Monitoring networks for air pollution have been established in numerous industrial zones. The data collected by the pervasive monitoring devices can be utilized not only for determining the current environmental condition, but also for forecasting it in the near future. This paper considers the applications of different machine learning methods for the prediction of the two most widely used quantities. Particulate matter (PM) with a diameter of 2.5 and 10 µm, respectively. The data are collected via a proprietary monitoring station, designated as the Ecolumn. The Ecolumn monitors a number of key parameters, including temperature, pressure, humidity, PM 1.0, PM 2.5, and PM 10, in a timely manner. The data were employed in the development of multiple models based on selected machine learning methods. The decision tree, random forest, recurrent neural network, and long short-term memory models were employed. Experiments were conducted with varying hyperparameters and network architectures. Different time scales (10 min, 1 h, and 24 h) were examined. The most optimal results were observed for the Long Short-Term Memory algorithm when utilizing the shortest available time spans (shortest averaging times). The decision tree and random forest algorithms demonstrated unexpectedly high performance for long averaging times, exhibiting only a slight decline in accuracy compared to neural networks for shorter averaging times. Recommendations for the potential applicability of the tested methods were formulated. [ABSTRACT FROM AUTHOR]

Purpose Methods Results Conclusion Level of Evidence The objective of this study is to train and test machine‐learning (ML) models to automatically classify shoulder rehabilitation exercises.The cohort included both healthy and patients with rotator‐cuff (RC) tears. All participants performed six shoulder rehabilitation exercises, following guidelines developed by the American Society of Shoulder and Elbow Therapists. Each exercise was repeated six times, while wearing a wearable system equipped with three magneto‐inertial sensors. Six supervised machine‐learning models (k‐Nearest Neighbours, Support Vector Machine, Decision Tree, Random Forest (RF), Logistic Regression and Adaptive Boosting) were trained for the classification. The algorithms' ability to accurately classify exercise activities was evaluated using the nested cross‐validation method, with different combinations of outer and inner folds.A total of 19 healthy subjects and 17 patients with complete RC tears were enroled in the study. The highest classification performances were achieved by the RF classifier, with an accuracy of 89.91% and an F1‐score of 89.89%.The results of this study highlight the feasibility and effectiveness of using wearable sensors and ML algorithms to accurately classify shoulder rehabilitation exercises. These findings suggest promising prospects for implementing the proposed wearable system in remote home‐based monitoring scenarios. The ease of setup and modularity of the system reduce user burden enabling patient‐driven sensor positioning.Level III. [ABSTRACT FROM AUTHOR]

The Colorado River has experienced a significant streamflow reduction in recent decades due to climate change, resulting in pronounced hydrological droughts that pose challenges to the environment and human activities. However, current models struggle to accurately capture complex drought patterns, and their accuracy decreases as the lead time increases. Thus, determining the reliability of drought forecasting for specific months ahead presents a challenging task. This study introduces a robust approach that utilizes the Beluga Whale Optimization (BWO) algorithm to train and optimize the parameters of the Regularized Extreme Learning Machine (RELM) and Random Forest (RF) models. The applied models are validated against a KNN benchmark model for forecasting drought from one- to six-month ahead across four hydrological stations distributed over the Colorado River. The achieved results demonstrate that RELM-BWO outperforms RF-BWO and KNN models, achieving the lowest root-mean square error (0.2795), uncertainty (U95 = 0.1077), mean absolute error (0.2104), and highest correlation coefficient (0.9135). Also, the current study uses Global Multi-Criteria Decision Analysis (GMCDA) as an evaluation metric to assess the reliability of the forecasting. The GMCDA results indicate that RELM-BWO provides reliable forecasts up to four months ahead. Overall, the research methodology is valuable for drought assessment and forecasting, enabling advanced early warning systems and effective drought countermeasures. [ABSTRACT FROM AUTHOR]

Soft robots have the advantage of adaptability and flexibility in various scenarios and tasks due to their inherent flexibility and mouldability, which makes them highly promising for real‐world applications. The development of electronic skin (E‐skin) perception systems is crucial for the advancement of soft robots. However, achieving both exteroceptive and proprioceptive capabilities in E‐skins, particularly in terms of decoupling and classifying sensing signals, remains a challenge. This study presents an E‐skin with mixed electronic and ionic conductivity that can simultaneously achieve exteroceptive and proprioceptive, based on the resistance response of conductive hydrogels. It is integrated with soft robots to enable state perception, with the sensed signals further decoded using the machine learning model of decision trees and random forest algorithms. The results demonstrate that the newly developed hydrogel sensing system can accurately predict attitude changes in soft robots when subjected to varying degrees of pressing, hot pressing, bending, twisting, and stretching. These findings that multifunctional hydrogels combine with machine learning to decode signals may serve as a basis for improving the sensing capabilities of intelligent soft robots in future advancements. [ABSTRACT FROM AUTHOR]

Life satisfaction is vital for older adults' well‐being, impacting various life aspects. It is dynamic, necessitating nuanced approaches to capture its trajectories accurately. This study aimed to explore the diverse trajectories and predictors of life satisfaction among older adults in China using longitudinal data from the China Health and Retirement Longitudinal Study. Latent class growth modeling and growth mixture modeling were employed to identify distinct trajectories of life satisfaction. Machine learning (ML) models were developed to predict different trajectories and identify important predictors of different trajectories. Four distinct trajectories of life satisfaction were identified, showcasing nuanced patterns of life satisfaction that changed over time. ML models, especially random forest, effectively predicted these trajectories. Emotional experiences (particularly the frequency of happiness and loneliness), body mass index, and self‐report health emerged as significant predictors of different life satisfaction trajectories. Our finding revealed the importance of focusing on individuals or groups with consistently low life satisfaction and paying more attention to mental and physical health predictors. Our models might guide future targeted preventative treatments. [ABSTRACT FROM AUTHOR]

This study delves into decoding hand gestures using surface electromyography (EMG) signals collected via a precision Myo-armband sensor, leveraging machine learning algorithms. The research entails rigorous data preprocessing to extract features and labels from raw EMG data. Following partitioning into training and testing sets, four traditional machine learning models are scrutinized for their efficacy in classifying finger movements across seven distinct gestures. The analysis includes meticulous parameter optimization and five-fold cross-validation to evaluate model performance. Among the models assessed, the Random Forest emerges as the top performer, consistently delivering superior precision, recall, and F1-score values across gesture classes, with ROC-AUC scores surpassing 99%. These findings underscore the Random Forest model as the optimal classifier for our EMG dataset, promising significant advancements in healthcare rehabilitation engineering and enhancing human–computer interaction technologies. [ABSTRACT FROM AUTHOR]

In the context of early disease detection, machine learning (ML) has emerged as a vital tool. Feature selection (FS) algorithms play a crucial role in ensuring the accuracy of predictive models by identifying the most influential variables. This study, focusing on a retrospective cohort of 4778 COVID-19 patients from Iran, explores the performance of various FS methods, including filter, embedded, and hybrid approaches, in predicting mortality outcomes. The researchers leveraged 115 routine clinical, laboratory, and demographic features and employed 13 ML models to assess the effectiveness of these FS methods based on classification accuracy, predictive accuracy, and statistical tests. The results indicate that a Hybrid Boruta-VI model combined with the Random Forest algorithm demonstrated superior performance, achieving an accuracy of 0.89, an F1 score of 0.76, and an AUC value of 0.95 on test data. Key variables identified as important predictors of adverse outcomes include age, oxygen saturation levels, albumin levels, neutrophil counts, platelet levels, and markers of kidney function. These findings highlight the potential of advanced FS techniques and ML models in enhancing early disease detection and informing clinical decision-making. [ABSTRACT FROM AUTHOR]

Machine learning methods have improved in recent years and provide increasingly powerful tools for understanding landscape evolution. In this study, we used the random forest method based on Google Earth Engine to evaluate the desertification dynamics in northern China from 1995 to 2020. We selected Landsat series image bands, remote sensing inversion data, climate baseline data, land use data, and soil type data as variables for majority voting in the random forest method. The method's average classification accuracy was 91.6% ± 5.8 [mean ± SD], and the average kappa coefficient was 0.68 ± 0.09, suggesting good classification results. The random forest classifier results were consistent with the results of visual interpretation for the spatial distribution of different levels of desertification. From 1995 to 2000, the area of aeolian desertification increased at an average rate of 9977 km2 yr−1, and from 2000 to 2005, from 2005 to 2010, from 2010 to 2015, and from 2015 to 2020, the aeolian desertification decreased at an average rate of 2535, 3462, 1487, and 4537 km2 yr−1, respectively. [ABSTRACT FROM AUTHOR]

Soil erodibility (K) refers to the inherent ability of soil to withstand erosion. Accurate estimation and spatial prediction of K values are vital for assessing soil erosion and managing land resources. However, as most K-value estimation models are empirical, they suffer from significant extrapolation uncertainty, and traditional studies on spatial prediction focusing on individual empirical K values have neglected to explore the spatial pattern differences between various empirical models. This work proposed a universal framework for selecting an optimal soil-erodibility map using empirical models enhanced by machine learning. Specifically, three empirical models, namely, the erosion-productivity impact calculator model (K_EPIC), the Shirazi model (K_Shirazi), and the Torri model (K_Torri) were used to estimate K values. Random Forest (RF) and Gradient-Boosting Decision Tree (GBDT) algorithms were employed to develop prediction models, which led to the creation of three K-value maps. The spatial distribution of K values and associated environmental covariates were also investigated across varying empirical models. Results showed that RF achieved the highest accuracy, with R2 of K_EPIC, K_Shirazi, and K_Torri increasing by 46%, 34%, and 22%, respectively, compared to GBDT. And distinctions among environmental variables that shape the spatial patterns of empirical models have been identified. The K_EPIC and K_Shirazi are influenced by soil porosity and soil moisture. The K_Torri is more sensitive to soil moisture conditions and terrain location. More importantly, our study has highlighted disparities in the spatial patterns across the three K-value maps. Considering the data distribution, spatial distribution, and measured K values, the K_Torri model outperformed others in estimating soil erodibility in the plateau lake watershed. This study proposed a framework that aimed to create optimal soil-erodibility maps and offered a scientific and accurate K-value estimation method for the assessment of soil erosion. [ABSTRACT FROM AUTHOR]

This paper focuses on large-scale landslide susceptibility modelling in NW Croatia. The objective of this research was to provide new insight into stable and unstable area sampling strategies on a representative inventory of small and shallow landslides mainly occurring in soil and soft rock. Four strategies were tested for stable area sampling (random points, stable area polygon, stable polygon buffering and stable area centroid) in combination with four strategies for unstable area sampling (landslide polygon, smoothing digital terrain model derived landslide conditioning factors, polygon buffering and landslide centroid), resulting in eight sampling scenarios. Using Logistic Regression, Neural Network, Random Forest and Support Vector Machine algorithm, 32 models were derived and analysed. The main conclusions reveal that polygon sampling of unstable areas is an imperative in large-scale modelling, as well as that subjective and/or biased stable area sampling leads to misleading models. Moreover, Random Forest and Neural Network proved to be more favourable methods (0.804 and 0.805 AUC, respectively), but also showed extreme sensitivity to the tested sampling strategies. In the comprehensive comparison, the advantages and disadvantages of 32 derived models were analysed through quantitative and qualitative parameters to highlight their application to large-scale landslide zonation. The results yielded by this research are beneficial to the susceptibility modelling step in large-scale landslide susceptibility assessments as they enable the derivation of more reliable zonation maps applicable to spatial and urban planning systems. [ABSTRACT FROM AUTHOR]

Brazil, a leading rice producer globally, faces challenges in systematically mapping its diverse rice fields due to varying cropping systems, climates, and planting calendars. Existing rice mapping methods often rely on complex techniques like deep learning or microwave imagery, posing limitations for large-scale mapping. This study proposes a novel approach utilizing Sentinel-2 spectral–temporal metrics (STMs) in conjunction with a random forest classifier for rice paddy mapping. By extracting diverse STMs and training both regional and global classifiers, we validated the method across independent areas. While regional models tended to overestimate rice areas, the global model effectively reduced discrepancies between our data and the reference maps, achieving an overall classifier accuracy exceeding 80%. Despite the need for further refinement to address confusion with other crops, STM exhibits promise for national-scale rice paddy mapping in Brazil. [ABSTRACT FROM AUTHOR]

Monitoring air pollution is important for human health and the environment. Previous studies on the prediction of air pollutants from satellite images have employed machine learning, yet there are few enhancements to the constructure of model. Moreover, the existing models have been successful in predicting pollutants like PM2.5, PM10, and O3. They have not been as effective in predicting other primary air pollutants. To improve the overall prediction performance of the existing model, a novel residual learning backpropagation model, abs. as BresNet, has been proposed in this research. This model has revealed the availability to precisely predict the ground-surface concentration of the six primary air pollutants, PM2.5, PM10, O3, NO2, CO, and SO2, based on the satellite imagery of MODIS AOD. Two of the most commonly used machine learning models so far, viz. the multilayer backpropagation neural network (MLBPN) and random forest (RF), were employed as the control. In the conducted experiments, the proposed BresNet model demonstrated significant improvements of 18.75%/31.94%, 33.82%/85.71%, 15.00%/35.29%, 39.06%/134.21%, 23.23%/68.00%, and 137.14%/260.87% in terms of R2 for the six primary air pollutants, compared to the RF/MLBPN model. Moreover, the convergence speed and loss function of the BresNet model compared to that of the MLBPN decreased by 55.15%, revealing superior convergence speed with the lower loss function. [ABSTRACT FROM AUTHOR]

Carbon dioxide corrosion is a pervasive issue in pipelines and the petroleum industry, posing substantial risks to equipment safety and longevity. Accurate prediction of corrosion rates and severity is essential for effective material selection and equipment maintenance. This paper begins by addressing the limitations of traditional corrosion prediction methods and explores the application of machine learning algorithms in C O 2 corrosion prediction. Conventional models often fail to capture the complex interactions among multiple factors, resulting in suboptimal prediction accuracy, limited adaptability, and poor generalization. To overcome these limitations, this study systematically organized and analyzed the data, performed a correlation analysis of the data features, and examined the factors influencing corrosion. Subsequently, prediction models were developed using six algorithms: Random Forest (RF), K-Nearest Neighbors (KNN), Gradient Boosting Decision Tree (GBDT), Support Vector Machine (SVM), XGBoost, and LightGBM. The results revealed that SVM exhibited the lowest performance on both training and test sets, while RF achieved the best results with R2 values of 0.92 for the training set and 0.88 for the test set. In the classification of corrosion severity, RF, LightGBM, SVM, and KNN were utilized, with RF demonstrating superior performance, achieving an accuracy of 99% and an F1-score of 0.99. This study highlights that machine learning algorithms, particularly Random Forest, offer substantial potential for predicting and classifying C O 2 corrosion. These algorithms provide innovative approaches and valuable insights for practical applications, enhancing predictive accuracy and operational efficiency in corrosion management. [ABSTRACT FROM AUTHOR]

The next generation of advanced materials is tending toward increasingly complex compositions. Synthesizing precise composition is time-consuming and becomes exponentially demanding with increasing compositional complexity. An experienced human operator does significantly better than a novice but still struggles to consistently achieve precision when synthesis parameters are coupled. The time to optimize synthesis becomes a barrier to exploring scientifically and technologically exciting compositionally complex materials. This investigation demonstrates an active learning (AL) approach for optimizing physical vapor deposition synthesis of thin-film alloys with up to five principal elements. We compared AL-based on Gaussian process (GP) and random forest (RF) models. The best performing models were able to discover synthesis parameters for a target quinary alloy in 14 iterations. We also demonstrate the capability of these models to be used in transfer learning tasks. RF and GP models trained on lower dimensional systems (i.e., ternary, quarternary) show an immediate improvement in prediction accuracy compared to models trained only on quinary samples. Furthermore, samples that only share a few elements in common with the target composition can be used for model pre-training. We believe that such AL approaches can be widely adapted to significantly accelerate the exploration of compositionally complex materials. [ABSTRACT FROM AUTHOR]

A variational heteroencoder based on recurrent neural networks, trained with SMILES linear notations of molecular structures, was used to derive the following atomic descriptors: delta latent space vectors (DLSVs) obtained from the original SMILES of the whole molecule and the SMILES of the same molecule with the target atom replaced. Different replacements were explored, namely, changing the atomic element, replacement with a character of the model vocabulary not used in the training set, or the removal of the target atom from the SMILES. Unsupervised mapping of the DLSV descriptors with t-distributed stochastic neighbor embedding (t-SNE) revealed a remarkable clustering according to the atomic element, hybridization, atomic type, and aromaticity. Atomic DLSV descriptors were used to train machine learning (ML) models to predict 19F NMR chemical shifts. An R2 of up to 0.89 and mean absolute errors of up to 5.5 ppm were obtained for an independent test set of 1046 molecules with random forests or a gradient-boosting regressor. Intermediate representations from a Transformer model yielded comparable results. Furthermore, DLSVs were applied as molecular operators in the latent space: the DLSV of a halogenation (H→F substitution) was summed to the LSVs of 4135 new molecules with no fluorine atom and decoded into SMILES, yielding 99% of valid SMILES, with 75% of the SMILES incorporating fluorine and 56% of the structures incorporating fluorine with no other structural change. [ABSTRACT FROM AUTHOR]

Background: Age-related Macular Degeneration (AMD) poses a growing global health concern as the leading cause of central vision loss in elderly people. Objection: This study focuses on unraveling the intricate involvement of Natural Killer (NK) cells in AMD, shedding light on their immune responses and cytokine regulatory roles. Methods: Transcriptomic data from the Gene Expression Omnibus database were utilized, employing single-cell RNA-seq analysis. High-dimensional weighted gene co-expression network analysis (hdWGCNA) and single-cell regulatory network inference and clustering (SCENIC) analysis were applied to reveal the regulatory mechanisms of NK cells in early-stage AMD patients. Machine learning models, such as random forests and decision trees, were employed to screen hub genes and key transcription factors (TFs) associated with AMD. Results: Distinct cell clusters were identified in the present study, especially the T/NK cluster, with a notable increase in NK cell abundance observed in AMD. Cell-cell communication analyses revealed altered interactions, particularly in NK cells, indicating their potential role in AMD pathogenesis. HdWGCNA highlighted the turquoise module, enriched in inflammation-related pathways, as significantly associated with AMD in NK cells. The SCENIC analysis identified key TFs in NK cell regulatory networks. The integration of hub genes and TFs identified CREM, FOXP1, IRF1, NFKB2, and USF2 as potential predictors for AMD through machine learning. Conclusion: This comprehensive approach enhances our understanding of NK cell dynamics, signaling alterations, and potential predictive models for AMD. The identified TFs provide new avenues for molecular interventions and highlight the intricate relationship between NK cells and AMD pathogenesis. Overall, this study contributes valuable insights for advancing our understanding and management of AMD. [ABSTRACT FROM AUTHOR]

• Machine learning (ML) techniques were pioneered to analyze non-linear relationships. • Algorithm's predictive power validated ML models, preventing spurious relations. • Recursive feature elimination explored predictor importance. • Presence of non-linear relationships is suggested, advising further investigation. • Acknowledgment of skepticism stresses rigorous analysis for long-term correlations. The Earth's climate is a complex system that can be influenced by various internal and external factors. Previous research has explored the relationship between climate and external forcings such as solar activity and GCR. However, the relations are intricate and intertwined into almost chaotic meteorological measurements. This paper delves deeper into understanding their interactions. GCR flux, Sunspot number (SSN), total solar irradiance (TSI), UV irradiance (UVI), and the Oceanic Niño Index (ONI) are the predictor variables, while total cloud amount (TCA) and low cloud amount (LCA) are the response variables. The analysis begins with standard statistical techniques and continues with multiple regression models including random forest, which is a machine learning (ML) method that can be used for non-linear regression. Subsequently, Recursive Feature Elimination (RFE) is employed to scrutinize the correlation among the predictor parameters. In the ML model, the final 25% of the dataset is held out and tested for validation, so, the predictive power of the algorithm is measured. Both geographical and temporal patterns have been investigated. This study suggests that a non-linear relationship might exist between the parameters, and should be investigated further, particularly in specific regions of the world. [ABSTRACT FROM AUTHOR]

This article investigates the behavior of hybrid FRP Concrete-Steel columns with an elliptical cross section. The investigation was carried out by gathering information through literature and conducting a parametric study, which resulted in 116 data points. Moreover, multiple machine learning predictive models were developed to accurately estimate the confined ultimate strain and the ultimate load of confined concrete at the rupture of FRP tube. Decision Tree (DT), Random Forest (RF), Adaptive Boosting (ADAB), Categorical Boosting (CATB), and eXtreme Gradient Boosting (XGB) machine learning techniques were utilized for the proposed models. Finally, these models were visually and quantitatively verified and evaluated. It was concluded that the CATB and XGB are standout models, offering high accuracy and strong generalization capabilities. The CATB model is slightly superior due to its consistently lower error rates during testing, indicating it is the best model for this dataset when considering both accuracy and robustness against overfitting. [ABSTRACT FROM AUTHOR]

Credit scorecards are essential tools for banks to assess the creditworthiness of loan applicants. While advanced machine learning models like XGBoost and random forest often outperform traditional logistic regression in predictive accuracy, their lack of interpretability hinders their adoption in practice. This study bridges the gap between research and practice by developing a novel framework for constructing interpretable credit scorecards using Shapley values. We apply this framework to two credit datasets, discretizing numerical variables and utilizing one-hot encoding to facilitate model development. Shapley values are then employed to derive credit scores for each predictor variable group in XGBoost, random forest, LightGBM, and CatBoost models. Our results demonstrate that this approach yields credit scorecards with interpretability comparable to logistic regression while maintaining superior predictive accuracy. This framework offers a practical and effective solution for credit practitioners seeking to leverage the power of advanced models without sacrificing transparency and regulatory compliance. [ABSTRACT FROM AUTHOR]

Depression, a prevalent global mental health disorder, often leads to a reduced quality of life and an increased risk of suicide. Despite the availability of treatments, many cases go undetected, highlighting the need for an accurate machine learning (ML) prediction model for depression severity and risk factors, particularly when dealing with multimodal datasets. Previous studies that utilized ML to predict the severity of depression encountered limitations, such as small datasets and a lack of personalization. This study proposes an optimal algorithm for predicting depression severity and personalized risk factors using ML. The potential bene- fits include improved accuracy in severity assessment, personalized treatment strategies, and refined risk factor identification. The random forest (RF) algorithm emerged as the most effective, exhibiting notable performance metrics on NHANES data, including a 0.93 R-squared, 0.93 explained variance score (EVS), 0.51 mean absolute error (MAE), 1.73 mean squared error (MSE), and 1.32 root mean squared error (RMSE). Notably, RF identified both general and personalized risk factors for depression severity. This model holds promise for clinical assessment, diagnosis, and intervention planning, contributing significantly to the comprehensive management of depression. [ABSTRACT FROM AUTHOR]

The current world has witnessed the emergence of various illnesses, such as autism spectrum disorder (ASD), that are not yet medically recognized. It impacts multiple behavioral domains, such as repetitive and stereotyped behavior, social competence, and linguistic skills. This condition is a severe neurodevelopmental disorder. It Identifying and classifying ASD is challenging and time-consuming due to its symptoms being remarkably similar to those of many other mental illnesses. Machine learning-based models are increasingly being used to predict a wide range of human diseases, leveraging various physiological and other characteristics. Our study aims to develop a classification model that can predict the likelihood of ASD in various age groups, such as toddlers, children, adolescents, and adults. We have utilized several machine learning (ML) algorithms, including support vector machine (SVM), Naive Bayes (NB), random forest (RF), extra trees classifier (ET), k-nearest neighbor (K-NN), decision tree (DT), Ada boost classifier (AB), and stochastic gradient descent (SGD) classifiers. These models are tested using four unique non-clinical ASD screening datasets that are publicly available from Kaggle and the UCI library. In the first dataset, there are 1054 instances and 19 features related to toddlers. The remaining ones consist of 21 traits and, for children, adolescents, and adults, 292, 104, and 704 cases, respectively. The outcomes of the experimentation have shown that the SDG, DT, and ET classifiers are the most commonly used models and have achieved results with almost 100% accuracy. [ABSTRACT FROM AUTHOR]

Breast cancer is currently one of the most prevalent cancers affecting women globally. Uncontrolled growth and division of breast cells lead to the formation of tumors, marking the onset of breast cancer. Predicting breast cancer is essential for early detection, making treatment plans, and implementing preventive measures, ultimately improving patient outcomes and reducing mortality rates. In recent years, numerous studies have been published to predict breast cancer where researchers use a variety of methods. Most investigations have been conducted using narrow and specific datasets, often resulting in a lack of accuracy. Such methods may not be suitable for clinical use. The study aims to address the limitations of existing models in terms of robustness and generalization across diverse datasets. In our study, we employed two metaheuristic algorithms, namely, genetic algorithm (GA) and chemical reaction optimization (CRO) with machine learning techniques, including support vector machine (SVM), decision tree, random forest, and XGBoost. GA and CRO are used to optimize the feature selection process. It enables machine learning algorithms to predict more accurately. Experiments were conducted on three datasets, namely, Wisconsin Breast Cancer (WBC), Breast Cancer‐the University of California, Irvine (BC‐UCI), and Breast Cancer Coimbra (BCC) datasets. The datasets contain 569, 286, and 116 instances, respectively. The classifiers with optimized features consistently outperformed the classifiers without feature optimization in terms of accuracy, precision, recall, specificity, and F1 score. Among the compared methods published recently, our method attained the highest accuracies of 99.64% in the WBC dataset and 98% in the BCC dataset, as well as the second highest accuracy of 99.12% in the BC‐UCI dataset. Comparative analysis demonstrated the superiority of our approach over existing methods. [ABSTRACT FROM AUTHOR]

Color characteristics are a crucial indicator of green tea quality, particularly in needle-shaped green tea, and are predominantly evaluated through subjective sensory analysis. Thus, the necessity arises for an objective, precise, and efficient assessment methodology. In this study, 885 images from 157 samples, obtained through computer vision technology, were used to predict sensory evaluation results based on the color features of the images. Three machine learning methods, Random Forest (RF), Support Vector Machine (SVM) and Decision Tree-based AdaBoost (DT-AdaBoost), were carried out to construct the color quality evaluation model. Notably, the DT-Adaboost model shows significant potential for application in evaluating tea quality, with a correct discrimination rate (CDR) of 98.50% and a relative percent deviation (RPD) of 14.827 in the 266 samples used to verify the accuracy of the model. This result indicates that the integration of computer vision with machine learning models presents an effective approach for assessing the color quality of needle-shaped green tea. [ABSTRACT FROM AUTHOR]

Background: Migraine is a common primary headache that has a significant impact on patients' quality of life. The co-occurrence of migraine and depression is frequent, resulting in more complex symptoms and a poorer prognosis. The evidence suggests that depression and migraine comorbidity share a polygenic genetic background. Objective: The aim of this study is to identify related genetic variants that contribute to genetic susceptibility to migraine with and without depression in a Chinese cohort. Methods: In this case-control study, 263 individuals with migraines and 223 race-matched controls were included. Eight genetic polymorphism loci selected from the GWAS were genotyped using Sequenom's MALDI-TOF iPLEX platform. Results: In univariate analysis, ANKDD1B rs904743 showed significant differences in genotype and allele distribution between migraineurs and controls. Furthermore, a machine learning approach was used to perform multivariate analysis. The results of the Random Forest algorithm indicated that ANKDD1B rs904743 was a significant risk factor for migraine susceptibility in China. Additionally, subgroup analysis by the Boruta algorithm showed a significant association between this SNP and migraine comorbid depression. Migraineurs with depression have been observed to have worse scores on the Beck Anxiety Inventory (BAI) and the Migraine Disability Assessment Scale (MIDAS). Conclusion: The study indicates that there is an association between ANKDD1B rs904743 and susceptibility to migraine with and without depression in Chinese patients. [ABSTRACT FROM AUTHOR]

Landslide disasters, due to their widespread distribution and clustered occurrences, pose a significant threat to human society. Rainfall is considered a primary triggering factor, and the frequent clustering of landslides underscores the importance of early warning systems for regional landslide disasters in preventing and mitigating rainfall-induced landslides. Research on early warning models is crucial for accurately predicting rainfall-induced landslides. However, traditional models face challenges such as the complexity of landslide causes, insufficient data, and limited analysis methods, resulting in low accuracy and inadequate precision. This study focuses on Fujian Province, China, proposing a four-step process for building a regional landslide early warning model based on machine learning. The process includes data integration and cleaning, sample set construction, model training and validation, and practical application. By integrating and cleaning the latest and most detailed data, a training sample set (15,589 samples) for the regional landslide disaster early warning model is established. Three machine learning algorithms--Random Forest, Multilayer Perceptron, and Convolutional Neural Network--are employed and compared, the evaluation results indicated that the RF-based warning model achieved an accuracy of 0.930-0.957 and an AUC value of 0.955. The CNNbased warning model demonstrated an accuracy of 0.945-0.948 with an AUC value of 0.940. The MLP-based warning model achieved an accuracy of 0.930-0.953 and an AUC value of 0.930. The results showed comparable accuracy metrics among the three models, with RF exhibiting a significant advantage in AUC values. Finally, the models are applied to the regional landslide disasters induced by heavy rainfall in Fujian Province on 5 August 2021. The results showed that in the binary classification warning strategy, the accuracy of the Random Forest and Convolutional Neural Network was 92.9%, while that of the Multilayer Perceptron was 85.8%, all performing well. In the multi-classification hierarchical warning strategy, the Random Forest excelled, while the performance of the Convolutional Neural Network and Multilayer Perceptron was relatively limited. The findings of this study Frontiers in Earth Science 01 frontiersin.org Liu et al. 10.3389/feart.2024.1419421 contribute to valuable attempts in landslide disaster warning model research, with anticipated further improvements through the gradual accumulation of samples and practical application verification. [ABSTRACT FROM AUTHOR]

Soil classification is essential for understanding soil properties and their suitability for conveying the characteristics of soil types. In this study, we present a prediction of soil classification using fewer soil variables by employing the random forest (RF) technique in machine learning. This study compiled the parameters outlined in the unified soil classification system (USCS), a widely used method for categorizing soils based on their properties and behavior. These parameters, encompassing grain size distribution, Atterberg limits, the coefficient of uniformity, and the coefficient of curvature, were defined within specific ranges to create a synthetic database for training the RF model. The importance of input factors in soil classification was assessed using the out-of-bag samples in RF. Through rigorous validation techniques, including cross-validation, the performance of the RF model is thoroughly assessed, demonstrating its capability to accurately evaluate soil classification. The findings indicate that the RF model presented in this study exhibits a promising alternative, providing automated and accurate classification based on soil data. Notably, the model indicates that the coefficients of uniformity and gradation are insignificant for soil classification and can predict soil types even when these factors are missing, a feat that traditional methods struggle to achieve. [ABSTRACT FROM AUTHOR]

Arsenic (As) contamination in groundwater represents a major global health threat, potentially impacting billions of individuals. Elevated As concentrations are found in river floodplains across south and southeast Asia, as well as in the inland basins of China, despite varying sedimentological and hydrogeochemical conditions. The specific mechanisms responsible for these high As levels remain poorly understood, complicating efforts to predict and manage the contamination. Applying hydro-chemical, geological, and soil parameters as explanatory variables, this study employs multiple linear regression (MLIR) and random forest regression (RFR) models to estimate groundwater As concentrations in these regions. Additionally, random forest classification (RFC) and multivariate logistic regression (MLOR) models are applied to predict the probability of As levels exceeding 10 μg/L in the Hetao Basin (China) and Bangladesh. Model validation reveals that RFR explains 80% and 70% of spatial variability of As concentration in the Hetao Basin and Bangladesh, respectively, outperforming MLIR, which accounts for only 35% and 32%. Similarly, RFC outperforms MLOR in predicting high As probability, achieving correct classification rates of 98.70% (Hetao Basin) and 98.25% (Bangladesh) on training datasets, and 82.76% (Hetao Basin) and 91.20% (Bangladesh) on validation datasets. The performance of the MLOR model on the validation set yields accuracy rates of 81.60% and 72.18%, respectively. In the Hetao Basin, Ca2+, redox potential (Eh), Fe, pH, SO42−, and Cl− are key predictors of As contamination, while in Bangladesh, soil organic carbon (SOC), pH, and SO42− are significant predictors. This study underscores the potential of random forest (RF) models as robust tools for predicting groundwater As contamination. [ABSTRACT FROM AUTHOR]

This study compares the calculation of snow water equivalent (SWE) using machine learning algorithms with the conventional degree-day method. The study uses machine learning techniques such as LASSO, Random Forest, Support Vector Machines, and CatBoost. It proposes an innovative use of feature engineering (FE) to improve the accuracy and robustness of SWE predictions by machine learning intended for interpolation, extrapolation, or imputation of missing data. The performance of machine learning approaches is evaluated against the traditional degree-day method for predicting SWE. The study emphasizes and demonstrates gains when modeling is enhanced by transforming basic, raw data through feature engineering. The results, verified in a case study from the mountainous region of Slovakia, suggest that machine learning, particularly CatBoost with feature engineering, shows better results in SWE estimation in comparison with the degree-day method, although the authors present a refined application of the degree-day method by utilizing genetic algorithms. Nevertheless, the study finds that the degree-day method achieved accuracy with a Nash–Sutcliffe coefficient of efficiency NSE = 0.59, while the CatBoost technique enhanced with the proposed FE achieved an accuracy NSE = 0.86. The results of this research contribute to refining snow hydrology modeling and optimizing SWE prediction for improved decision-making in snow-dominated regions. [ABSTRACT FROM AUTHOR]

Climate change is making water management increasingly difficult due to rising temperatures and unpredictable rainfall patterns, impacting crop water availability and irrigation needs. This study investigated the ability of machine learning and satellite remote sensing to monitor water status and physiology. The research focused on predicting different eco-physiological parameters in an irrigated peach orchard under Mediterranean conditions, utilizing multispectral reflectance data and machine learning algorithms (extreme gradient boosting, random forest, support vector regressor); ground data were acquired from 2021 to 2023 in the south of Italy. The random forest model outperformed in predicting net assimilation (R2 = 0.61), while the support vector machine performed best in predicting electron transport rate (R2 = 0.57), Fv/Fm ratio (R2 = 0.66) and stomatal conductance (R2 = 0.56). Random forest also proved to be the most effective in predicting stem water potential (R2 = 0.62). These findings highlighted the potential of integrating machine learning techniques with high-resolution satellite imagery to assist farmers in monitoring crop health and optimizing irrigation practices, thereby addressing the challenges determined by climate change. [ABSTRACT FROM AUTHOR]

A major psychological problem that numerous new mothers experience is postpartum depression (PPD). A woman's capacity to care for herself and her child may be hampered by the sadness, anxiety, and weariness it can bring on. Using information from their past medical treatment, lifestyles, and habits, deep learning can assist in identifying women who are at risk for PPD. By analyzing large amounts of data from various sources, deep learning models can help healthcare providers provide targeted support and interventions to women most likely to develop PPD. This can improve outcomes for both mothers and babies and help to reduce the overall burden of PPD on families and society. To identify and validate risk variables for PPD disease in women at an early stage, this investigation provides a CNN-BLSTM with a TL-based model. The hybrid model combines transfer learning (TL) with bi-directional long short-term memory (BLSTM) models and convolutional neural networks (CNNs) models. This model takes input data such as medical history, demographic information, and behavioural patterns and outputs the probability of the patient having PPD. The CNN component of the model mainly extracts the relevant features from the input data, while the LSTM component helps to analyze the sequential patterns in the data. By combining the strengths of both CNN and BLSTM, the model can effectively capture complex relationships within the input data and make accurate predictions. In the proposed model, we used it for transfer learning after Fine-tuning a pre-trained CNN + BLSTM model on a large dataset of non-PPD data to extract features. TL method helps to enhance the accuracy. We utilize an online PPD dataset which contains 1503 records of various women. The proposed model and well-known ML methods, i.e., Decision Tree, Random Forest, AdaBoost, LightGBM, XGBoost and CatBoost, were compared using various performance measuring parameters, i.e., Precision, Recall, F1-score, accuracy, and ROC_AUC. The proposed hybrid model achieved 99.6% accuracy and 99.7% precision which is higher than other methods. [ABSTRACT FROM AUTHOR]

Objective: Accurate prediction of survival prognosis is helpful to guide clinical decision-making. The aim of this study was to develop a model using machine learning techniques to predict the occurrence of composite thromboembolic events (CTEs) in elderly patients with atrial fibrillation(AF). These events encompass newly diagnosed cerebral ischemia events, cardiovascular events, pulmonary embolism, and lower extremity arterial embolism. Methods: This retrospective study included 6,079 elderly hospitalized patients (≥ 75 years old) with AF admitted to the People's Liberation Army General Hospital in China from January 2010 to June 2022. Random forest imputation was used for handling missing data. In the descriptive statistics section, patients were divided into two groups based on the occurrence of CTEs, and differences between the two groups were analyzed using chi-square tests for categorical variables and rank-sum tests for continuous variables. In the machine learning section, the patients were randomly divided into a training dataset (n = 4,225) and a validation dataset (n = 1,824) in a 7:3 ratio. Four machine learning models (logistic regression, decision tree, random forest, XGBoost) were trained on the training dataset and validated on the validation dataset. Results: The incidence of composite thromboembolic events was 19.53%. The Least Absolute Shrinkage and Selection Operator (LASSO) method, using 5-fold cross-validation, was applied to the training dataset and identified a total of 18 features that exhibited a significant association with the occurrence of CTEs. The random forest model outperformed other models in terms of area under the curve (ACC: 0.9144, SEN: 0.7725, SPE: 0.9489, AUC: 0.927, 95% CI: 0.9105–0.9443). The random forest model also showed good clinical validity based on the clinical decision curve. The Shapley Additive exPlanations (SHAP) showed that the top five features associated with the model were history of ischemic stroke, high triglyceride (TG), high total cholesterol (TC), high plasma D-dimer, age. Conclusions: This study proposes an accurate model to stratify patients with a high risk of CTEs. The random forest model has good performance. History of ischemic stroke, age, high TG, high TC and high plasma D-Dimer may be correlated with CTEs. [ABSTRACT FROM AUTHOR]

Floods are among the natural hazards that have seen a rapid increase in frequency in recent decades. The damage caused by floods, including human and financial losses, poses a serious threat to human life. This study evaluates two machine learning (ML) techniques for flood susceptibility mapping (FSM) in the Gamasyab watershed in Iran. We utilized random forest (RF), support vector machine (SVM), ensemble models, and a geographic information system (GIS) to predict FSM. The application of these models involved 10 effective factors in flooding, as well as 82 flood locations integrated into the GIS. The SVM and RF models were trained and tested, followed by the implementation of resampling techniques (RT) using bootstrap and subsampling methods in three repetitions. The results highlighted the importance of elevation, slope, and precipitation as primary factors influencing flood occurrence. Additionally, the ensemble model outperformed both the RF and SVM models, achieving an area under the curve (AUC) of 0.9, a correlation coefficient (COR) of 0.79, a true skill statistic (TSS) of 0.83, and a standard deviation (SD) of 0.71 in the test phase. The tested models were adapted to available input data to map the FSM across the study watershed. These findings underscore the potential of integrating an ensemble model with GIS as an effective tool for flood susceptibility mapping. [ABSTRACT FROM AUTHOR]

As artificial intelligence (AI) becomes widespread, there is increasing attention on investigating bias in machine learning (ML) models. Previous research concentrated on classification problems, with little emphasis on regression models. This paper presents an easy-to-apply and effective methodology for mitigating bias in bagging and boosting regression models, that is also applicable to any model trained through minimizing a differentiable loss function. Our methodology measures bias rigorously and extends the ML model's loss function with a regularization term to penalize high correlations between model errors and protected attributes. We applied our approach to three popular tree-based ensemble models: a random forest model (RF), a gradient-boosted model (GBT), and an extreme gradient boosting model (XGBoost). We implemented our methodology on a case study for predicting road-level traffic volume, where RF, GBT, and XGBoost models were shown to have high accuracy. Despite high accuracy, the ML models were shown to perform poorly on roads in minority-populated areas. Our bias mitigation approach reduced minority-related bias by over 50%. [ABSTRACT FROM AUTHOR]

Objective: To develop a machine learning-based risk prediction model for postoperative parastomal hernia (PSH) in colorectal cancer patients undergoing permanent colostomy, assisting nurses in identifying high-risk groups and devising preventive care strategies. Methods: A case-control study was conducted on 495 colorectal cancer patients who underwent permanent colostomy at the Second Affiliated Hospital of Anhui Medical University from June 2017 to June 2023, with a 1-year follow-up period. Patients were categorized into PSH and non-PSH groups based on PSH occurrence within 1-year post-operation. Data were split into training (70%) and testing (30%) sets. Variable selection was performed using Least Absolute Shrinkage and Selection Operator (LASSO) regression, and binary classification prediction models were established using Logistic Regression (LR), Support Vector Classification (SVC), K Nearest Neighbor (KNN), Random Forest (RF), Light Gradient Boosting Machine (LGBM), and Extreme Gradient Boosting (XgBoost). The binary classification label denoted 1 for PSH occurrence and 0 for no PSH occurrence. Parameters were optimized via 5-fold cross-validation. Model performance was evaluated using Area Under Curve (AUC), specificity, sensitivity, accuracy, positive predictive value, negative predictive value, and F1-score. Clinical utility was evaluated using decision curve analysis (DCA), model explanation was enhanced using shapley additive explanation (SHAP), and model visualization was achieved using a nomogram. Results: The incidence of PSH within 1 year was 29.1% (144 patients). Among the models tested, the RF model demonstrated the highest discrimination capability with an AUC of 0.888 (95% CI: 0.881–0.935), along with superior specificity, accuracy, sensitivity, and F1 score. It also showed the highest clinical net benefit on the DCA curve. SHAP analysis identified the top 10 influential variables associated with PSH risk: body mass index (BMI), operation duration, history and status of chronic obstructive pulmonary disease (COPD), prealbumin, tumor node metastasis (TNM) staging, stoma site, thickness of rectus abdominis muscle (TRAM), C-reactive protein CRP, american society of anesthesiologists physical status classification (ASA), and stoma diameter. These insights from SHAP plots illustrated how these factors influence individual PSH outcomes. The nomogram was used for model visualization. Conclusion: The Random Forest model demonstrated robust predictive performance and clinical relevance in forecasting colonic PSH. This model aids in early identification of high-risk patients and guides preventive care. [ABSTRACT FROM AUTHOR]

This study employs machine learning algorithms to analyze the bead foam extrusion process and to assess the impact of processing parameters, specifically focusing on their effects on bead foam density and melt pressure in under water granulation (UWG) for polylactic acid (PLA). These interrelated parameters, influenced by processing parameters such as temperature, screw speed, and blowing agent, possess challenges for traditional empirical methods to capture. The key factors that significantly impact the prediction of melt pressure in UWG are blowing agent, injector pressure, temperature in B‐extruder and die size. Likewise, essential parameters for predicting bead foam density comprise blowing agent, injector pressure, temperature in B‐extruder, die plate temperature, melt temperature in B‐extruder, and melt pressure in B‐extruder. Machine learning (ML) models were employed to forecast bead foam density and melt pressure in UWG using various processing parameters in PLA bead foam extrusion. The random forest model achieved a high coefficient of determination R2 score of 0.96 for predicting melt pressure in UWG. Additionally, the decision tree model demonstrated effective predictions for bead density, with the R2 score: 0.81. These ML models can be applied to diverse materials, leading to more sustainable, efficient processes for bead foam extrusion. [ABSTRACT FROM AUTHOR]

Predicting phase stability in high entropy alloys (HEAs), such as phase fractions as functions of composition and temperature, is essential for understanding alloy properties and screening desirable materials. Traditional methods like CALPHAD are computationally intensive for exploring high-dimensional compositional spaces. To address such a challenge, this study explored and compared the effectiveness of random forests (RF) and deep neural networks (DNN) for accelerating materials discovery by building surrogate models of phase stability prediction. For interpolation scenarios (testing on the same order of system as trained), RF models generally produce smaller errors than DNN models. However, for extrapolation scenarios (training on lower-order systems and testing on higher order systems), DNNs generalize more effectively than traditional ML models. DNN demonstrate the potential to predict topologically relevant phase composition when data were missing, making it a powerful predictive tool in materials discovery frameworks. The study uses a CALPHAD dataset of 480 million data points generated from a custom database, available for further model development and benchmarking. Experiments show that DNN models are data-efficient, achieving similar performance with a fraction of the dataset. This work highlights the potential of DNNs in materials discovery, providing a powerful tool for predicting phase stability in HEAs, particularly within the Cr-Hf-Mo-Nb-Ta-Ti-V-W-Zr composition space. [ABSTRACT FROM AUTHOR]

Background: Depression is one of the most common mental illnesses among middle-aged and older adults in China. It is of great importance to find the crucial factors that lead to depression and to effectively control and reduce the risk of depression. Currently, there are limited methods available to accurately predict the risk of depression and identify the crucial factors that influence it. Methods: We collected data from 25,586 samples from the harmonized China Health and Retirement Longitudinal Study (CHARLS), and the latest records from 2018 were included in the current cross-sectional analysis. Ninety-three input variables in the survey were considered as potential influential features. Five machine learning (ML) models were utilized, including CatBoost and eXtreme Gradient Boosting (XGBoost), Gradient Boosting decision tree (GBDT), Random Forest (RF), Light Gradient Boosting Machine (LightGBM). The models were compared to the traditional multivariable Linear Regression (LR) model. Simultaneously, SHapley Additive exPlanations (SHAP) were used to identify key influencing factors at the global level and explain individual heterogeneity through instance-level analysis. To explore how different factors are nonlinearly associated with the risk of depression, we employed the Accumulated Local Effects (ALE) approach to analyze the identified critical variables while controlling other covariates. Results: CatBoost outperformed other machine learning models in terms of MAE, MSE, MedAE, and R2metrics. The top three crucial factors identified by the SHAP were r4satlife, r4slfmem, and r4shlta, representing life satisfaction, selfreported memory, and health status levels, respectively. Conclusion: This study demonstrates that the CatBoost model is an appropriate choice for predicting depression among middle-aged and older adults in Harmonized CHARLS. The SHAP and ALE interpretable methods have identified crucial factors and the nonlinear relationship with depression, which require the attention of domain experts. [ABSTRACT FROM AUTHOR]

Background: Nosocomial infections (NIs) frequently occur and adversely impact prognosis for hospitalized patients with cirrhosis. This study aims to develop and validate two machine learning models for NIs and in-hospital mortality risk prediction. Methods: The Prediction of Nosocomial Infection and Prognosis in Cirrhotic patients (PIPC) study included hospitalized patients with cirrhosis at the Qingchun Campus of the First Affiliated Hospital of Zhejiang University. We then assessed several machine learning algorithms to construct predictive models for NIs and prognosis. We validated the best-performing models with bootstrapping techniques and an external validation dataset. The accuracy of the predictions was evaluated through sensitivity, specificity, predictive values, and likelihood ratios, while predictive robustness was examined through subgroup analyses and comparisons between models. Results: We enrolled 1,297 patients into derivation cohort and 496 patients into external validation cohort. Among the six algorithms assessed, the Random Forest algorithm performed best. For NIs, the PIPC-NI model achieved an area under the curve (AUC) of 0.784 (95% confidence interval [CI] 0.741–0.826), a sensitivity of 0.712, and a specificity of 0.702. For in-hospital mortality, the PIPC- mortality model achieved an AUC of 0.793 (95% CI 0.749–0.836), a sensitivity of 0.769, and a specificity of 0.701. Moreover, our PIPC models demonstrated superior predictive performance compared to the existing MELD, MELD-Na, and Child-Pugh scores. Conclusions: The PIPC models showed good predictive power and may facilitate healthcare providers in easily assessing the risk of NIs and prognosis among hospitalized patients with cirrhosis. [ABSTRACT FROM AUTHOR]

This study explored the use of forearm electromyography data to distinguish eight hand gestures. The neural network (NN) and random forest (RF) algorithms were tested on data from 10 participants. As window sizes increase from 200 ms to 1000 ms, the algorithm accuracies increased with RF from 85% to 97% due to the increased temporal resolution. It was also noticed that the RF performed better with an accuracy of 85% than the NN with accuracy 80% when the temporal resolution was smaller, indicating the RF will be efficient when quick-response time is important. As the window size increases, the NN showed higher performance, suggesting that NN will be useful when higher accuracy is required. Future studies should increase the sample size, include more hand gestures, use different feature extraction methods and test different algorithms to improve the accuracy and efficiency of the system. [ABSTRACT FROM AUTHOR]

Low muscle mass is associated with numerous adverse outcomes independent of other associated comorbid diseases. We aimed to predict and understand an individual's risk for developing low muscle mass using proteomics and machine learning. We identified eight biomarkers associated with low pectoralis muscle area (PMA). We built three random forest classification models that used either clinical measures, feature selected biomarkers, or both to predict development of low PMA. The area under the receiver operating characteristic curve for each model was: clinical-only = 0.646, biomarker-only = 0.740, and combined = 0.744. We displayed the heterogenetic nature of an individual's risk for developing low PMA and identified two distinct subtypes of participants who developed low PMA. While additional validation is required, our methods for identifying and understanding individual and group risk for low muscle mass could be used to enable developments in the personalized prevention of low muscle mass. [ABSTRACT FROM AUTHOR]

AbstractPer- and polyfluoroalkyl substances (PFASs), one of the persistent organic pollutants, have immunosuppressive effects. The evaluation of this effect has been the focus of regulatory toxicology. In this investigation, 146 PFASs (immunosuppressive or nonimmunosuppressive) and corresponding concentration gradients were collected from literature, and their structures were characterized by using Dragon descriptors. Feature importance analysis and stepwise feature elimination are used for feature selection. Three machine learning (ML) methods, namely Random Forest (RF), Extreme Gradient Boosting Machine (XGB), and Categorical Boosting Machine (CB), were utilized for model development. The model interpretability was explored by feature importance analysis and correlation analysis. The findings indicated that the three models developed have exhibited excellent performance. Among them, the best-performing RF model has an average AUC score of 0.9720 for the testing set. The results of the feature importance analysis demonstrated that concentration, SpPosA_X, IVDE, R2s, and SIC2 were the crucial molecular features. Applicability domain analysis was also performed to determine reliable prediction boundaries for the model. In conclusion, this study is the first application of ML models to investigate the immunosuppressive activity of PFASs. The variables used in the models can help understand the mechanism of the immunosuppressive activity of PFASs, allow researchers to more effectively assess the immunosuppressive potential of a large number of PFASs, and thus better guide environmental and health risk assessment efforts. [ABSTRACT FROM AUTHOR]

Introduction Methods Results Conclusion The platelet fluorescent counting (PLT‐F) method is utilized as a reflex test method following the initial test of the platelet impedance counting (PLT‐I) method in clinical practice on the Sysmex XN‐series automated hematology analyzer. Our aim is to establish reflex test rules for the PLT‐F method by combining multiple parameters provided by the “CBC + DIFF” mode of the Sysmex XN‐series automated hematology analyzer.We tested 120 samples to evaluate the baseline bias between the PLT‐F and PLT‐I methods. Then, we selected 1256 samples to establish and test reflex test rules using seven machine learning models (decision Tree, random forest, neural network, logistic regression, k‐nearest neighbor, support vector machine, and Naive Bayes). The training set and test set were divided at a ratio of 7:3. We evaluated the performance of machine learning models on the test set using various metrics to select the most valuable model.The PLT‐F method exhibited a high degree of correlation with the PLT‐I method (r = 0.998). The random forest model emerged as the most valuable, boasting an accuracy of 0.893, an area under the curve of 0.954, an F1 score of 0.771, a recall of 0.719, a precision of 0.831, and a specificity of 0.950. The most important variable in the random forest model was mean cell volume, weighted at 15.09%.The random forest model, which demonstrated high efficiency in our study, can be used to establish PLT reflex test rules based on the PLT‐F method for the Sysmex XN‐series automated hematology analyzer. [ABSTRACT FROM AUTHOR]