Search

Machine learning classifiers are increasingly widely used. This research note explores how a particular widely used classifier, the Random Forest, performs when faced with samples which are imbalanced and noisy data. Both are known to affect accuracy, but if their effects are independent or not has not been explored. Based on an experiment using synthetic data generated for the study we find that the effects of noise and sample balance interact with each other; classification accuracy is worse when faced with both noisy data and sample imbalance. This has implications for the use of RF in market research, but also for how methods to address either sample imbalance or noise are assessed. [ABSTRACT FROM AUTHOR]

This study developed an innovative biosensor strategy for the sensitive and selective detection of canine mammary tumor biomarkers, cancer antigen 15–3 (CA 15–3) and mucin 1 (MUC-1), integrating green silver nanoparticles (GAgNPs) with machine learning (ML) algorithms to achieve high diagnostic accuracy and potential for noninvasive early detection. The GAgNPs-enhanced electrochemical biosensor demonstrated selective detection of CA 15–3 in serum and MUC-1 in tissue homogenates, with limits of detection (LODs) of 0.07 and 0.11 U mL−1, respectively. The nanoscale dimensions of the GAgNPs endowed them with electrochemically active surface areas, facilitating sensitive biomarker detection. Experimental studies targeted CA 15–3 and MUC-1 biomarkers in clinical samples, and the biosensor exhibited ease of use and good selectivity. Furthermore, ML algorithms were employed to analyze the electrochemical data and predict biomarker concentrations, enhancing the diagnostic accuracy. The Random Forest algorithm achieved 98% accuracy in tumor presence prediction, while an Artificial Neural Network attained 76% accuracy in CA 15–3-based tumor grade classification. The integration of ML techniques with the GAgNPs-based biosensor offers a promising approach for noninvasive, accurate, and early detection of canine mammary tumors, potentially revolutionizing veterinary diagnostics. This multilayered strategy, combining eco-friendly nanomaterials, electrochemical sensing, and ML algorithms, holds significant potential for advancing both biomedical research and clinical practice in the field of canine mammary tumor diagnostics. [ABSTRACT FROM AUTHOR]

The internet of things (IoT) has seen rapid expansion, but this growth brings significant privacy challenges due to the large amounts of data generated by myriad IoT devices. To address these challenges, this study introduces Smart Anonymity, a method that determines the optimal data anonymization algorithm for a dataset by assessing its unique features. The solution leverages OWL ontologies grounded in description logic (DL), which facilitates inconsistency checks and the discovery of new facts for data validation. Additionally, machine learning (ML) is incorporated to improve the accuracy of these classifications. ML is also instrumental in recommending suitable anonymization algorithms, with the random forest algorithm being employed explicitly for this purpose. The findings from this research indicate that Smart Anonymity effectively improves user privacy. [ABSTRACT FROM AUTHOR]

For a long time, decision-making in auditing was limited to a risk-oriented recommendation and consisted of the rigorous analysis of a sample of data. The new trend in the audit decision process focuses on the use of decision support systems (DSSs) founded on data analytics (DA) to better concentrate on in-depth analysis. This study examines how DA can improve the audit decision-making approach in the banking sector. We show that DA techniques can improve the quality of audit decision-making within banks and highlight the advantages associated with mastering these techniques, which results in a more effective and efficient audit of digital banking transactions. We propose an artifact-based data analytics-driven decision support system (DA-DSS) for an automatic fraud detection system supported by DA. The proposed DA-DSS artifact with a random forest classifier at its core is a promising innovation in the field of electronic transaction fraud detection. The results show that the random forest classifier can accurately classify the data generated by this artifact with an accuracy varying from 88 to 93% using transaction data collected from 2021 to 2022. Other classifiers including k-nearest neighbors (KNN) are also used, with a classification rate ranging from 66 to 88% for the same transaction datasets. These results show that the proposed DA-DSS with random forest can significantly improve auditing by reducing the time required for data analysis and increasing the results' accuracy. Management can use the proposed artifact to enhance and speed up the decision-making process within their organization. [ABSTRACT FROM AUTHOR]

High costs and time consuming associated with experimental trial-and-error result in low efficiency, creating an urgent need for a more effective strategy for ultra-high temperature ceramic matrix composites (UHTCMCs) development. Inspired by the exceptional performance of machine learning (ML) algorithms across various domains, this work employs ML algorithms to construct models and conduct in-depth analysis of the key factors and their patterns influencing the ablation resistance of UHTCMCs. A set of 26 dimensional features that could potentially impact the ablation resistance of UHTCMCs were established based on domain knowledge. Eight typical ML models were used to build and predict the linear ablation rate (LAR) of UHTCMCs. Results show that the random forest model has optimal generalization performance, with mean absolute error (MAE), mean squared error (MSE), and coefficient of determination (R 2) being 2.75 μm s−1 and 7.3 (μm s−1)2, and 0.71 respectively. The Shapley additive explanations values based on the random forest model reveal that the key features affecting the LAR of UHTCMCs are ranked as average melting point of ceramics (AMPC) > thermal conductivity of material (TCM) > thermal expansion coefficient of oxides (TECO) > fabrication temperature of material (FTM), all showing a negative influence on the LAR. Symbolic regression further indicates that AMPC, TCM, and TECO have an exponential negative correlation with LAR. These data-driven conclusions have been thoroughly validated through the use of C f /(TiZrHfNbTa)C composites. The established model can accelerate the discovery of material knowledge and provide reliable guidance for UHTCMC development. [ABSTRACT FROM AUTHOR]

Metamaterial absorbers (MMAs) have tremendous potential for controlling and modulating Terahertz electromagnetic (EM) waves. It is challenging to design MMAs for optimal performance using conventional methods due to their time-consuming and computationally demanding nature. Machine learning (ML) regressor approaches have emerged as potential tools for optimizing the predictive modeling of MMAs in recent years. This article examined different regression techniques, such as Decision Tree, k-Nearest Neighbors, Random Forest, Extra Trees (ET), Extreme Gradient Boosting, Bagging, and Categorical Boosting. The primary goal is to assess the effectiveness of each regressor technique in forecasting the performance of MMAs using various performance metrics. The study focuses on a representative spectrum of MMAs, chosen for its key features that are common across a wide range of MMAs. The extra tree Regressor outperforms other models in the comparative analysis due to its capacity to integrate multiple decision trees and include randomness into the feature selection process, resulting in improved predictive skills. The ET regressor forecasts MMA performance remarkably well for a 70 % test size, with a Root Mean Squared Error of 0.0251, a Mean Absolute Error of 0.0074, a Mean Squared Error of 0.006, and an Adjusted R-squared of 0.9873. These findings can assist researchers in optimizing metamaterial absorber designs for Terahertz applications using ML regressor techniques and provide insights into how these methods can be generalized to other spectra with similar characteristics. [ABSTRACT FROM AUTHOR]

Motivation: The association between weather conditions and stroke incidence has been a subject of interest for several years, yet the findings from various studies remain inconsistent. Additionally, predictive modelling in this context has been infrequent. This study explores the relationship of extremely high ischaemic stroke incidence and meteorological factors within the Slovak population. Furthermore, it aims to construct forecasting models of extremely high number of strokes. Methods: Over a five-year period, a total of 52,036 cases of ischemic stroke were documented. Days exhibiting a notable surge in ischemic stroke occurrences (surpassing the 90th percentile of historical records) were identified as extreme cases. These cases were then scrutinized alongside daily meteorological parameters spanning from 2015 to 2019. To create forecasts for the occurrence of these extreme cases one day in advance, three distinct methods were employed: Logistic regression, Random Forest for Time Series, and Croston's method. Results: For each of the analyzed stroke centers, the cross-correlations between instances of extremely high stroke numbers and meteorological factors yielded negligible results. Predictive performance achieved by forecasts generated through multivariate logistic regression and Random Forest for time series analysis, which incorporated meteorological data, was on par with that of Croston's method. Notably, Croston's method relies solely on the stroke time series data. All three forecasting methods exhibited limited predictive accuracy. Conclusions: The task of predicting days characterized by an exceptionally high number of strokes proved to be challenging across all three explored methods. The inclusion of meteorological parameters did not yield substantive improvements in forecasting accuracy. [ABSTRACT FROM AUTHOR]

In an era marked by pervasive digital connectivity, cybersecurity concerns have escalated. The rapid evolution of technology has led to a spectrum of cyber threats, including sophisticated zero-day attacks. This research addresses the challenge of existing intrusion detection systems in identifying zero-day attacks using the CIC-MalMem-2022 dataset and autoencoders for anomaly detection. The trained autoencoder is integrated with XGBoost and Random Forest, resulting in the models XGBoost-AE and Random Forest-AE. The study demonstrates that incorporating an anomaly detector into traditional models significantly enhances performance. The Random Forest-AE model achieved 100% accuracy, precision, recall, F1 score, and Matthews Correlation Coefficient (MCC), outperforming the methods proposed by Balasubramanian et al., Khan, Mezina et al., Smith et al., and Dener et al. When tested on unseen data, the Random Forest-AE model achieved an accuracy of 99.9892%, precision of 100%, recall of 99.9803%, F1 score of 99.9901%, and MCC of 99.8313%. This research highlights the effectiveness of the proposed model in maintaining high accuracy even with previously unseen data. [ABSTRACT FROM AUTHOR]

Background: To design a pulmonary ground-glass nodules (GGN) classification method based on computed tomography (CT) radiomics and machine learning for prediction of invasion in early-stage ground-glass opacity (GGO) pulmonary adenocarcinoma. Methods: This retrospective study included pulmonary GGN patients who were histologically confirmed to have adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma cancer (IAC) from 2020 to 2023. CT images of all patients were automatically segmented and 107 radiomic features were obtained for each patient. Classification models were developed using random forest (RF) and cross-validation, including three one-versus-others models and one three-class model. For each model, features were ranked by normalized Gini importance, and a minimal subset was selected with a cumulative importance exceeding 0.9. These selected features were then used to train the final models. The models' performance metrics, including area under the curve (AUC), accuracy, sensitivity, and specificity, were computed. AUC and accuracy were compared to determine the final optimal method. Results: The study comprised 193 patients (mean age 54 ± 11 years, 65 men), including 65 AIS, 54 MIA, and 74 IAC, divided into one training cohort (N = 154) and one test cohort (N = 39). The final three-class RF model outperformed three individual one-versus-others models in distinguishing each class from the other two. For the multiclass classification model, the AUC, accuracy, sensitivity, and specificity were 0.87, 0.79, 0.62, and 0.88 for AIS; 0.90, 0.79, 0.54, and 0.89 for MIA; and 0.87, 0.69, 0.73, and 0.67 for IAC, respectively. Conclusions: A radiomics-based multiclass RF model could effectively differentiate three types of pulmonary GGN, which enabled early diagnosis of GGO pulmonary adenocarcinoma. [ABSTRACT FROM AUTHOR]

Objective: This study aimed to develop a deep learning model to predict skeletal malocclusions with an acceptable level of accuracy using airway and cephalometric landmark values obtained from analyzing different CBCT images. Background: In orthodontics, multitudinous studies have reported the correlation between orthodontic treatment and changes in the anatomy as well as the functioning of the airway. Typically, the values obtained from various measurements of cephalometric landmarks are used to determine skeletal class based on the interpretation an orthodontist experiences, which sometimes may not be accurate. Methods: Samples of skeletal anatomical data were retrospectively obtained and recorded in Digital Imaging and Communications in Medicine (DICOM) file format. The DICOM files were used to reconstruct 3D models using 3DSlicer (slicer.org) by thresholding airway regions to build up 3D polygon models of airway regions for each sample. The 3D models were measured for different landmarks that included measurements across the nasopharynx, the oropharynx, and the hypopharynx. Male and female subjects were combined as one data set to develop supervised learning models. These measurements were utilized to build 7 artificial intelligence-based supervised learning models. Results: The supervised learning model with the best accuracy was Random Forest, with a value of 0.74. All the other models were lower in terms of their accuracy. The recall scores for Class I, II, and III malocclusions were 0.71, 0.69, and 0.77, respectively, which represented the total number of actual positive cases predicted correctly, making the sensitivity of the model high. Conclusion: In this study, it is observed that the Random Forest model was the most accurate model for predicting the skeletal malocclusion based on various airway and cephalometric landmarks. [ABSTRACT FROM AUTHOR]

Background and purpose: Early detection, diagnosis, and treatment of colorectal cancer and its precancerous lesions can significantly improve patients' survival rates. The purpose of this research is to identify methylation markers specific to colorectal cancer tissues and validate their diagnostic capability in colorectal cancer and precancerous changes by measuring the level of DNA methylation in stool samples. Method: We analyzed samples from six cancer tissues and adjacent normal tissues and fecal samples from 758 participants, including 62 patients with interfering diseases. Bioinformatics databases were used to screen for candidate biomarkers for CRC, and quantitative methylation-specific PCR methods were applied for identification. The methylation levels of the candidate biomarkers in fecal and tissue samples were measured. Logistic regression and random forest models were built and validated using fecal sample data from one of the centers, and the independent or combined diagnostic value of the candidate biomarkers in fecal samples for CRC and precancerous lesions was analyzed. Finally, the diagnostic capability and stability of the model were validated at another medical center. Results: This study identified two colorectal cancer CpG sites with tissue specificity. These two biomarkers have certain diagnostic power when used individually, but their diagnostic value for colorectal cancer and colorectal adenoma is more significant when they are used in combination. Conclusion: The results indicate that a DNA methylation biomarker combined diagnostic model based on two CpG sites, cg13096260 and cg12587766, has the potential for screening and diagnosing precancerous lesions and colorectal cancer. Additionally, compared to traditional diagnostic models, machine learning algorithms perform better but may yield more false-positive results, necessitating further investigation. [ABSTRACT FROM AUTHOR]

Background: Sepsis poses a critical threat to hospitalized patients, particularly those in the Intensive Care Unit (ICU). Rapid identification of Sepsis is crucial for improving survival rates. Machine learning techniques offer advantages over traditional methods for predicting outcomes. This study aimed to develop a prognostic model using a Stacking-based Meta-Classifier to predict 30-day mortality risks in Sepsis-3 patients from the MIMIC-III database. Methods: A cohort of 4,240 Sepsis-3 patients was analyzed, with 783 experiencing 30-day mortality and 3,457 surviving. Fifteen biomarkers were selected using feature ranking methods, including Extreme Gradient Boosting (XGBoost), Random Forest, and Extra Tree, and the Logistic Regression (LR) model was used to assess their individual predictability with a fivefold cross-validation approach for the validation of the prediction. The dataset was balanced using the SMOTE-TOMEK LINK technique, and a stacking-based meta-classifier was used for 30-day mortality prediction. The SHapley Additive explanations analysis was performed to explain the model's prediction. Results: Using the LR classifier, the model achieved an area under the curve or AUC score of 0.99. A nomogram provided clinical insights into the biomarkers' significance. The stacked meta-learner, LR classifier exhibited the best performance with 95.52% accuracy, 95.79% precision, 95.52% recall, 93.65% specificity, and a 95.60% F1-score. Conclusions: In conjunction with the nomogram, the proposed stacking classifier model effectively predicted 30-day mortality in Sepsis patients. This approach holds promise for early intervention and improved outcomes in treating Sepsis cases. [ABSTRACT FROM AUTHOR]

Bridge construction collapse is one of the most common bridge safety accidents. At present, evaluation results are often affected by the ability and experience of the assessor. Therefore, it is difficult to quickly, accurately and effectively evaluate the risk in the process of bridge construction. Moreover, key factors that can prevent accidents can hardly find from the existing bridge construction safety management and evaluation method. This paper analyzes and classifies the artificial and environmental risk factors that affect the bridge construction stage, and establishes 26 risk factors in 5 categories according to the characteristics of bridge construction and the actual situation of the project. Random forest (RF) algorithm is a non-parametric machine learning method based on decision tree, which does not need to be scored by experts in advance and avoids the influence of subjective factors. Compared with other analysis methods, random forest algorithm has the advantages of accurate and robust risk assessment results. Based on the advantages of random forest algorithm and the characteristics of bridge construction risk, this paper uses random forest algorithm to evaluate the bridge construction risk, and ranks the importance of indicators, and identify the index that has a greater influence on the risk. In order to verify the applicability and feasibility of the proposed method, a typical urban complex pedestrian bridge was taken as an example for actual engineering evaluation and verification. The results obtained are basically consistent with the actual risk assessment results of the pedestrian bridge. [ABSTRACT FROM AUTHOR]

Alpine natural heritage sites hold significant value due to their unique global resources. Studying land cover changes in these areas is crucial for maintaining and preserving multiple their values. This study takes Kalajun-Kuerdening, one of the components of Xinjiang Tianshan World Natural Heritage Site, as an example to analyze land cover changes and their driving factors in alpine heritage sites. Highlights include: (1) Between 1994 and 2023, Forest and Grassland increased by 55.96 km2 and 18.16 km2, with notable forest growth from 2007 to 2017. Trends in Forest changes align with forest protection policies, and a substantial amount of Bareland converted to Grassland indicates an increase in vegetation cover. (2) Elevation, precipitation, temperature, and evapotranspiration are key drivers of land cover changes, as validated by Random Forest algorithm and Geodetector model. (3) Favorable conditions for Grassland to Forest transition include annual precipitation between 275 and 375 mm, annual temperature between −2 and 3 °C, annual evapotranspiration between 580 and 750 mm, elevation between 1800 and 2600 m, and aspect between 0 to 110° and 220 to 259.9°. Continuous monitoring of land cover changes and their driving factors in mountain heritage sites contributes to the protection of the ecological environment and provides data and information support for addressing climate change, resource management, and policy making. [ABSTRACT FROM AUTHOR]

When conducting condition recognition research on AC contactor vibration signals through time-frequency analysis, the feature data exhibit a high degree of redundancy, which leads to repetitive information and hinders the accuracy of recognition. To address the redundancy issue in the features of AC contactor vibration signals, this study introduces a feature selection method based on Regularized Random Forest with Recursive Selection (RFRS). Initially, a test platform for AC contactor vibration signals was established, and time-frequency domain features of the AC contactor vibration signals were extracted. Subsequently, the traditional Random Forest (RF) was refined by optimizing its stopping criteria using the Recursive Feature Elimination approach and by incorporating a regularization coefficient during the splitting process to direct the split towards significant features. This modification not only enhances the Random Forest's capacity to leverage existing information but also introduces a bias, enabling it to favor important features. Finally, through case analysis, the proposed method effectively reduced the dimensionality of the feature set and achieved an average of 87.37% for Recall, 87.41% for F1-Score, 88.38% for Precision, and 85.74% for Accuracy. The overall performance of this method surpasses that of the three mainstream feature selection methods: Spearman's rank correlation coefficient method, the embedded method, and the filter method. This study thus provides a rather effective feature selection approach for the state recognition study of AC contactors. [ABSTRACT FROM AUTHOR]

In intensive care unit (ICU) patients undergoing mechanical ventilation (MV), the occurrence of difficult weaning contributes to increased ventilator-related complications, prolonged hospitalization duration, and a significant rise in healthcare costs. Therefore, early identification of influencing factors and prediction of patients at risk of difficult weaning can facilitate early intervention and preventive measures. This study aimed to strengthen airway management for ICU patients by constructing a risk prediction model with comprehensive and individualized offline programs based on machine learning techniques. This study involved the collection of data from 487 patients undergoing MV in the ICU, with a total of 36 variables recorded. The dataset was divided into a training set (70% of the data) and a test set (30% of the data). Five machine learning models, namely logistic regression, random forest, support vector machine, light gradient boosting machine, and extreme gradient boosting, were compared to predict the risk of difficult weaning in ICU patients with MV. Significant influencing factors were identified based on the results of these models, and a risk prediction model for ICU patients with MV was established. When evaluating the models using AUC (Area under the Curve of ROC) and Accuracy as performance metrics, the Random Forest algorithm exhibited the best performance among the five machine learning algorithms. The area under the operating characteristic curve for the subjects was 0.805, with an accuracy of 0.748, recall (0.888), specificity (0.767) and F1 score (0.825). This study successfully developed a risk prediction model for ICU patients with MV using a machine learning algorithm. The Random Forest algorithm demonstrated the highest prediction performance. These findings can assist clinicians in accurately assessing the risk of difficult weaning in patients and formulating effective individualized treatment plans. Ultimately, this can help reduce the risk of difficult weaning and improve the quality of life for patients. [ABSTRACT FROM AUTHOR]

Metabolic syndrome (MetS) is a cluster of interconnected metabolic risk factors, including abdominal obesity, high blood pressure, and elevated fasting blood glucose levels, that result in an increased risk of heart disease and stroke. In this research, we aim to identify the risk factors that have an impact on MetS in the Bangladeshi population. Subsequently, we intend to construct predictive machine learning (ML) models and ultimately, assess the accuracy and reliability of these models. In this particular study, we utilized the ATP III criteria as the basis for evaluating various health parameters from a dataset comprising 8185 participants in Bangladesh. After employing multiple ML algorithms, we identified that 27.8% of the population exhibited a prevalence of MetS. The prevalence of MetS was higher among females, accounting for 58.3% of the cases, compared to males with a prevalence of 41.7%. Initially, we identified the crucial variables using Chi-Square and Random Forest techniques. Subsequently, the obtained optimal variables are employed to train various models including Decision Trees, Random Forests, Support Vector Machines, Extreme Gradient Boosting, K-nearest neighbors, and Logistic Regression. Particularly we employed the ATP III criteria, which utilizes the Waist-to-Height Ratio (WHtR) as an anthropometric index for diagnosing abdominal obesity. Our analysis indicated that Age, SBP, WHtR, FBG, WC, DBP, marital status, HC, TGs, and smoking emerged as the most significant factors when using Chi-Square and Random Forest analyses. However, further investigation is necessary to evaluate its precision as a classification tool and to improve the accuracy of all classifiers for MetS prediction. [ABSTRACT FROM AUTHOR]

Background: Clear cell renal cell carcinoma (ccRCC) remains one of the most lethal urological malignancies even though a great number of improvements in diagnosis and management have achieved over the past few decades. Accumulated evidence revealed that histone deacetylases (HDACs) play vital role in cell proliferation, differentiation and apoptosis. Nevertheless, the biological functions of histone deacetylation modification related genes in ccRCC remains poorly understood. Method: Bulk transcriptomic data and clinical information of ccRCC patients were obtained from the TCGA database and collected from the Chinese PLA General Hospital. A total of 36 histone deacetylation genes were selected and studied in our research. Univariate cox regression analysis, least absolute shrinkage and selection operator (LASSO) regression, random forest (RF) analysis, and protein-protein interaction (PPI) network analysis were applied to identify key genes affecting the prognosis of ccRCC. The 'oncoPredict' algorithm was utilized for drug-sensitive analysis. Gene Set Enrichment Analysis (GSEA) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was used to explore the potential biological function. The ssGSEA algorithm was used for tumor immune microenvironment analysis. The expression levels of HDAC10 were validated by RT-PCR and immunohistochemistry (IHC). 5-ethynyl-2′-deoxyuridine (EdU assay), CCK-8 assay, cell transwell migration and invasion assay and colony formation assay were performed to detect the proliferation and invasion ability of ccRCC cells. A nomogram incorporating HDAC10 and clinicopathological characteristics was established to predict the prognosis of ccRCC patients. Result: Two machine learning algorithms and PPI analysis identified four histone deacetylation genes that have a significant association with the prognosis of ccRCC, with HDAC10 being the key gene among them. HDAC10 is highly expressed in ccRCC and its high expression is associated with poor prognosis for ccRCC patients. Pathway enrichment and the experiments of EdU staining, CCK-8 assay, cell transwell migration and invasion assay and colony formation assay demonstrated that HDAC10 mediated the proliferation and metastasis of ccRCC cells and involved in reshaping the tumor microenvironment (TME) of ccRCC. A clinically reliable prognostic predictive model was established by incorporating HDAC10 and other clinicopathological characteristics (https://nomogramhdac10.shinyapps.io/HDAC10%5fNomogram/). Conclusion: Our study found the increased expression of HDAC10 was closely associated with poor prognosis of ccRCC patients. HDAC10 showed a pro-tumorigenic effect on ccRCC and promote the proliferation and metastasis of ccRCC, which may provide new light on targeted therapy for ccRCC. [ABSTRACT FROM AUTHOR]

Objective: The aim of this study is to assess the use of machine learning methodologies in the diagnosis of endometriosis (EM). Methods: This study included a total of 106 patients with EM and 203 patients with non-EM conditions (like simple cysts and simple uterine fibroids), all admitted to the Shunyi Women's and Children's Hospital of Beijing Children's Hospital between January 2017 and September 2022. All participants were free of comorbidities and their diagnoses were confirmed via postoperative pathology. Comparative analysis was conducted between the EM and non-EM groups. Baseline data were assessed, including white blood cell count, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio, lymphocyte-to-monocyte ratio, mean platelet volume, hemoglobin, carbohydrate antigen 125 (CA125), carbohydrate antigen 199, coagulation parameters, and other serologic indicators. An optimal predictive model was developed using an artificial intelligence algorithm to determine the presence of EM. The objective is to provide new insights for the clinical diagnosis and treatment of EM. Results: The random forest algorithm demonstrated superior performance when compared to decision trees, LogitBoost, artificial neural networks, naïve Bayes, support vector machines, and linear regression in machine learning methods. Combining CA125 with the NLR yielded a better prediction of EM than using CA125 alone when applying the random forest algorithm. The accuracy of predicting EM with CA125 combined with NLR was 78.16%, with a sensitivity of 86.21% and an area under the curve (AUC) of 0.85 (P < 0.05). In contrast, using CA125 alone resulted in an EM prediction accuracy of 75.8%, with a sensitivity of 79.3% and an AUC of 0.82 (P < 0.05). Conclusion: The diagnostic value of serum CA125 combined with the NLR for EM is higher than that of serum CA125 alone. This finding indicates that NLR could serve as a new supplementary biomarker along with serum CA125 in the diagnosis of EM. [ABSTRACT FROM AUTHOR]

Many species of fishes around the world are soniferous. The types of sounds fishes produce vary among species and regions but consist typically of low-frequency (< 1.5 kHz) pulses and grunts. These sounds can potentially be used to monitor fishes non-intrusively and could complement traditional monitoring techniques. However, the significant time required for human analysts to manually label fish sounds in acoustic recordings does not yet allow passive acoustics to be used as a viable tool for monitoring fishes. In this paper, we compare two different approaches to automatically detect fish sounds. One is a more traditional machine learning technique based on the detection of acoustic transients in the spectrogram and the classification using Random Forest (RF). The other is using a deep learning approach and is based on the classification of overlapping segments (0.2 s) of spectrogram using a ResNet18 Convolutional Neural Network (CNN). Both algorithms were trained using 21,950 manually annotated fish and non-fish sounds collected from 2014 to 2019 at five different locations in the Strait of Georgia, British Columbia, Canada. The performance of the detectors was tested on part of the data from the Strait of Georgia that was withheld from the training phase, data from Barkley Sound, British Columbia, and data collected in the Port of Miami, Florida, United States. The CNN performed up to 1.9 times better than the RF (F1 score: 0.82 vs. 0.43). In some cases, the CNN was able to find more faint fish sounds than the analyst and performed well in environments different from the one it was trained in (Miami F1 score: 0.88). Noise analysis in the 20--1,000 Hz frequency band shows that the CNN is still reliable in noise levels greater than 130 dB re 1 μPa in the Port of Miami but becomes less reliable in Barkley Sound past 100 dB re 1 μPa due to mooring noise. The proposed approach can efficiently monitor (unidentified) fish sounds in a variety of environments and can also facilitate the development of species-specific detectors. We provide the software FishSound Finder, an easy-to-use open-source implementation of the CNN detector with detailed documentation. [ABSTRACT FROM AUTHOR]

This study examines the impact of internet usage frequency and behavior on social status discordance using CSS2021 data, and adopts a causal random forest model to test the heterogeneity of the impact of internet usage on social status discordance. The results reveal a clear inconsistency between objective and subjective social status among Chinese residents. In addition, increasing internet usage frequency can make actors more inclined to underestimate their social status. Moreover, capital enhancement through the internet significantly reduces the likelihood of status inflation. Finally, individuals with higher social status are less likely to have an upward bias in status cognition than those with lower social status when they use the internet frequently and use it for capital enhancement. These findings advance our understanding of social status identification among Chinese residents and provide different perspectives for studying social stratification in the digital age. [ABSTRACT FROM AUTHOR]

Motivation In the realm of precision medicine, effective patient stratification and disease subtyping demand innovative methodologies tailored for multi-omics data. Clustering techniques applied to multi-omics data have become instrumental in identifying distinct subgroups of patients, enabling a finer-grained understanding of disease variability. Meanwhile, clinical datasets are often small and must be aggregated from multiple hospitals. Online data sharing, however, is seen as a significant challenge due to privacy concerns, potentially impeding big data's role in medical advancements using machine learning. This work establishes a powerful framework for advancing precision medicine through unsupervised random forest-based clustering in combination with federated computing. Results We introduce a novel multi-omics clustering approach utilizing unsupervised random forests. The unsupervised nature of the random forest enables the determination of cluster-specific feature importance, unraveling key molecular contributors to distinct patient groups. Our methodology is designed for federated execution, a crucial aspect in the medical domain where privacy concerns are paramount. We have validated our approach on machine learning benchmark datasets as well as on cancer data from The Cancer Genome Atlas. Our method is competitive with the state-of-the-art in terms of disease subtyping, but at the same time substantially improves the cluster interpretability. Experiments indicate that local clustering performance can be improved through federated computing. Availability and implementation The proposed methods are available as an R-package (https://github.com/pievos101/uRF) [ABSTRACT FROM AUTHOR]

Objective: Data imbalance is a pervasive issue in medical data mining, often leading to biased and unreliable predictive models. This study aims to address the urgent need for effective strategies to mitigate the impact of data imbalance on classification models. We focus on quantifying the effects of different imbalance degrees and sample sizes on model performance, identifying optimal cut-off values, and evaluating the efficacy of various methods to enhance model accuracy in highly imbalanced and small sample size scenarios. Methods: We collected medical records of patients receiving assisted reproductive treatment in a reproductive medicine center. Random forest was used to screen the key variables for the prediction target. Various datasets with different imbalance degrees and sample sizes were constructed to compare the classification performance of logistic regression models. Metrics such as AUC, G-mean, F1-Score, Accuracy, Recall, and Precision were used for evaluation. Four imbalance treatment methods (SMOTE, ADASYN, OSS, and CNN) were applied to datasets with low positive rates and small sample sizes to assess their effectiveness. Results: The logistic model's performance was low when the positive rate was below 10% but stabilized beyond this threshold. Similarly, sample sizes below 1200 yielded poor results, with improvement seen above this threshold. For robustness, the optimal cut-offs for positive rate and sample size were identified as 15% and 1500, respectively. SMOTE and ADASYN oversampling significantly improved classification performance in datasets with low positive rates and small sample sizes. Conclusions: The study identifies a positive rate of 15% and a sample size of 1500 as optimal cut-offs for stable logistic model performance. For datasets with low positive rates and small sample sizes, SMOTE and ADASYN are recommended to improve balance and model accuracy. [ABSTRACT FROM AUTHOR]

Traditionally, the design of an expert system involves acquiring knowledge, in the form of symbolic rules, directly from the expert(s), which is a complex and time‐consuming task. Although expert systems approach is quite old, it is still present, especially where explicit knowledge representation and reasoning, which assure interpretability and explainability, are necessary. Therefore, machine learning methods have been devised to extract rules from data, to facilitate that task. However, those methods are quite inflexible in adapting to the application domain and provide no help in designing the expert system. In this work, we present a framework and corresponding tool, namely ACRES, for semi‐automatically generating expert systems from datasets. ACRES allows for data preprocessing, which helps in structuring knowledge in the form of a tree, called rule hierarchy, which represents (possible) dependencies among data variables and is used for rule formation. This improves interpretability and explainability of the produced systems. We have also designed and evaluated alternative methods for rule extraction from data and for calculation and use of certainty factors, to represent uncertainty; CFs can be dynamically updated. Experimental results on seven well‐known datasets show that the proposed rule extraction methods are comparable to other popular machine learning approaches like decision trees, CART, JRip, PART, Random Forest, and so on, for the classification task. Finally, we give insights on two applications of ACRES. [ABSTRACT FROM AUTHOR]

Combining different drugs synergistically is an essential aspect of developing effective treatments. Although there is a plethora of research on computational prediction for new combination therapies, there is limited to no research on combination therapies in the treatment of viral diseases. This paper proposes AI-based models for predicting novel antiviral combinations to treat virus diseases synergistically. To do this, we assembled a comprehensive dataset comprising information on viral strains, drug compounds, and their known interactions. As far as we know, this is the first dataset and learning model on combination therapy for viruses. Our proposal includes using a random forest model, an SVM model, and a deep model to train viral combination therapy. The machine learning models showed the highest performance, and the predicted values were validated by a t-test, indicating the effectiveness of the proposed methods. One of the predicted combinations of acyclovir and ribavirin has been experimentally confirmed to have a synergistic antiviral effect against herpes simplex type-1 virus, as described in the literature. [ABSTRACT FROM AUTHOR]

The chloride migration coefficient (CMC) of concrete is crucial for evaluating its durability. This study develops ensemble models to predict the CMC of concrete, addressing the limitations of traditional, labor-intensive laboratory tests. We developed three ensemble models: an inverse variance-based model, an Artificial Neural Network (ANN)-based model, and a tree-based model using the random forest regression algorithm. These models were trained on a dataset comprising 843 concrete mix proportions from existing literature. Results indicate that ensemble models outperform single models such as ANN and Support Vector Regression (SVR) in predicting CMC, with the combined random forest and ANN model showing the highest accuracy. Sensitivity analysis using Shapley Additive Explanations (SHAP) reveals that the CMC is most influenced by the water-to-cement ratio and curing age. Additionally, we designed a graphical user interface (GUI) to facilitate the practical application of our models. This research offers a robust methodology for evaluating concrete durability and potential for extending the prediction to other concrete properties. [ABSTRACT FROM AUTHOR]

Groundwater is a primary source of drinking water for billions worldwide. It plays a crucial role in irrigation, domestic, and industrial uses, and significantly contributes to drought resilience in various regions. However, excessive groundwater discharge has left many areas vulnerable to potable water shortages. Therefore, assessing groundwater potential zones (GWPZ) is essential for implementing sustainable management practices to ensure the availability of groundwater for present and future generations. This study aims to delineate areas with high groundwater potential in the Bankura district of West Bengal using four machine learning methods: Random Forest (RF), Adaptive Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost), and Voting Ensemble (VE). The models used 161 data points, comprising 70% of the training dataset, to identify significant correlations between the presence and absence of groundwater in the region. Among the methods, Random Forest (RF) and Extreme Gradient Boosting (XGBoost) proved to be the most effective in mapping groundwater potential, suggesting their applicability in other regions with similar hydrogeological conditions. The performance metrics for RF are very good with a precision of 0.919, recall of 0.971, F1-score of 0.944, and accuracy of 0.943. This indicates a strong capability to accurately predict groundwater zones with minimal false positives and negatives. Adaptive Boosting (AdaBoost) demonstrated comparable performance across all metrics (precision: 0.919, recall: 0.971, F1-score: 0.944, accuracy: 0.943), highlighting its effectiveness in predicting groundwater potential areas accurately; whereas, Extreme Gradient Boosting (XGBoost) outperformed the other models slightly, with higher values in all metrics: precision (0.944), recall (0.971), F1-score (0.958), and accuracy (0.957), suggesting a more refined model performance. The Voting Ensemble (VE) approach also showed enhanced performance, mirroring XGBoost's metrics (precision: 0.944, recall: 0.971, F1-score: 0.958, accuracy: 0.957). This indicates that combining the strengths of individual models leads to better predictions. The groundwater potentiality zoning across the Bankura district varied significantly, with areas of very low potentiality accounting for 41.81% and very high potentiality at 24.35%. The uncertainty in predictions ranged from 0.0 to 0.75 across the study area, reflecting the variability in groundwater availability and the need for targeted management strategies. In summary, this study highlights the critical need for assessing and managing groundwater resources effectively using advanced machine learning techniques. The findings provide a foundation for better groundwater management practices, ensuring sustainable use and conservation in Bankura district and beyond. [ABSTRACT FROM AUTHOR]

The mechanisms underlying severe allergic asthma are complex and unknown, meaning it is a challenge to provide the most appropriate treatment. This study aimed to identify novel biomarkers for stratifying allergic asthmatic patients according to severity, and to uncover the biological mechanisms that lead to the development of the severe uncontrolled phenotype. By using miRNA PCR panels, we analyzed the expression of 752 miRNAs in serum samples from control subjects (n = 15) and mild (n = 11) and severe uncontrolled (n = 10) allergic asthmatic patients. We identified 40 differentially expressed miRNAs between severe uncontrolled and mild allergic asthmatic patients. Functional enrichment analysis revealed signatures related to inflammation, angiogenesis, lipid metabolism and mRNA regulation. A random forest classifier trained with DE miRNAs achieved a high accuracy of 97% for severe uncontrolled patient stratification. Validation of the identified biomarkers was performed on a subset of allergic asthmatic patients from the CAMP cohort at Brigham and Women's Hospital, Harvard Medical School. Four of these miRNAs (hsa-miR-99b-5p, hsa-miR-451a, hsa-miR-326 and hsa-miR-505-3p) were validated, pointing towards their potential as biomarkers for stratifying allergic asthmatic patients by severity and providing insights into severe uncontrolled asthma molecular pathways. [ABSTRACT FROM AUTHOR]

Spherical data arise widely in various settings. Spherical statistics is an analysis of data on a unit hyper-spherical domain. In this paper, we mainly consider the local kernel estimators for regression models with a binary response and the predictors including spherical variables. We apply the random forests kernel to nonparametric binary regression models with spherical predictors. Simulation experiments and real examples are used to validate the performance of the new models. Compared with the classical von Mises–Fisher kernel and the linear-spherical kernel, the random forests kernel has better fitting effect and faster computation speed. Compared with other classifiers, the models proposed in this paper have better classification performance in both low and high dimensional cases. [ABSTRACT FROM AUTHOR]

Visual information affects static postural control, but how it affects dynamic postural control still needs to be fully understood. This study investigated the effect of proprioception weighting, influenced by the presence or absence of visual information, on dynamic posture control during voluntary trunk movements. We recorded trunk movement angle and angular velocity, center of pressure (COP), electromyographic, and electroencephalography signals from 35 healthy young adults performing a standing trunk flexion–extension task under two conditions (Vision and No-Vision). A random forest analysis identified the 10 most important variables for classifying the conditions, followed by a Wilcoxon signed-rank test. The results showed lower maximum forward COP displacement and trunk flexion angle, and faster maximum flexion angular velocity in the No-Vision condition. Additionally, the alpha/beta ratio of the POz during the switch phase was higher in the No-Vision condition. These findings suggest that visual deprivation affects cognitive- and sensory-integration-related brain regions during movement phases, indicating that sensory re-weighting due to visual deprivation impacts motor control. The effects of visual deprivation on motor control may be used for evaluation and therapeutic interventions in the future. [ABSTRACT FROM AUTHOR]

The mechanical properties of fissured sandstone will deteriorate under water–rock interaction. It is crucial to extract the precursor information of fissured sandstone instability under water–rock interaction. The potential of each acoustic emission (AE) parameter as a precursor for instability in the failure process of fissured sandstone was investigated in this study. An experimental dataset comprising 586 acoustic emission experiments was established, and subsequent classification training and testing were conducted using three machine learning (ML) models: AdaBoost, MLP, and Random Forest (RF). The primary parameters for identifying the instability risk state of fissured sandstone include acoustic emission ringing count, energy (mV·ms), centroid frequency, peak frequency, Rise Angle (RA), Average Frequency (AF), b value, and the natural/saturated state of fissured sandstone: state. To enhance data utilization, a 10-fold cross-validation method was employed during the model training process. The machine learning models were developed and designed to identify the instability risk of fissured sandstone under the natural and saturated states. The results demonstrated that the established RF model was capable of identifying fissured sandstone instability risks with an accuracy of 97.87%. Feature importance analysis revealed that state and b value exerted the most significant influence on identification results. The Spearman correlation coefficient was utilized to assess the correlation between input features. This study can provide technical support to identify the risk of instability of fissured sandstones under both natural and saturated water conditions. Based on the models developed in this study, it is possible to implement an early warning method for instability in fissured sandstone that meets realistic working conditions. Compared with the traditional empirical and formulaic methods, the machine learning method can more quickly process huge amounts of AE data and accurately identify the damage state of fissured sandstone. [ABSTRACT FROM AUTHOR]

The study comprehensively evaluates low-cost CO2 sensors from different price tiers, assessing their performance against a reference-grade instrument and exploring the possibility of calibration using different machine learning techniques. Three sensors (Sunrise AB by Senseair, K30 CO2 by Senseair, and GMP 343 by Vaisala) were tested alongside a reference instrument (Los Gatos precision greenhouse gas analyzer). The results revealed differences in sensor performance, with the higher cost Vaisala sensors exhibiting superior accuracy. Despite its lower price, the Sunrise sensors still demonstrated reasonable accuracy. Meanwhile, the K30 sensor measurements displayed higher variability and noise. Machine learning models, including linear regression, gradient boosting regression, and random forest regression, were employed for sensor calibration. In general, linear regression models performed best for extrapolating data, whereas decision tree-based models were generally more useful in handling non-linear datasets. Notably, a stack ensemble model combining these techniques outperformed the individual models and significantly improved sensor accuracy by approximately 65%. Overall, this study contributes to filling the gap in intercomparing CO2 sensors across different price categories and underscores the potential of machine learning for enhancing sensor accuracy, particularly in low-cost sensor applications. [ABSTRACT FROM AUTHOR]

A common challenge for exoskeleton control is discerning operator intent to provide seamless actuation of the device with the operator. One way to accomplish this is with joint angle estimation algorithms and multiple sensors on the human–machine system. However, the question remains of what can be accomplished with just one sensor. The objective of this study was to deploy a modular testing approach to test the performance of two joint angle estimation models—a kinematic extrapolation algorithm and a Random Forest machine learning algorithm—when each was informed solely with kinematic gait data from a single potentiometer on an ankle exoskeleton mock-up. This study demonstrates (i) the feasibility of implementing a modular approach to exoskeleton mock-up evaluation to promote continuity between testing configurations and (ii) that a Random Forest algorithm yielded lower realized errors of estimated joint angles and a decreased actuation time than the kinematic model when deployed on the physical device. [ABSTRACT FROM AUTHOR]

Accurate estimation of the distribution of POC in the sea surface is an important issue in understanding the carbon cycle at the basin scale in the ocean. This study explores the best machine learning approach to determine the distribution of POC in the ocean surface layer based on data obtained using satellite remote sensing. In order to estimate and verify the accuracy of this method, it is necessary to obtain a large amount of POC data from field observations, so this study was conducted in the Mediterranean Sea, where such data have been obtained and published. The research initially utilizes the Geographic Detector (GD) method to identify spatial correlations between POC and 47 environmental factors in the region. Four machine learning models of a Bayesian optimized random forest (BRF), a backpropagation neural network, adaptive boosting, and extreme gradient boosting were utilized to construct POC assessment models. Model validation yielded that the BRF exhibited superior performance in estimating sea-surface POC. To build a more accurate tuneRanger random forest (TRRF) model, we introduced the tuneRanger R package for further optimization, resulting in an R2 of 0.868, a mean squared error of 1.119 (mg/m3)2, and a mean absolute error of 1.041 mg/m3. It was employed to estimate the surface POC concentrations in the Mediterranean for May and June 2017. Spatial analysis revealed higher concentrations in the west and north and lower concentrations in the east and south, with higher levels near the coast and lower levels far from the coast. Additionally, we deliberated on the impact of human activities on the surface POC in the Mediterranean. This research contributes a high-precision method for satellite retrieval of surface POC concentrations in the Mediterranean, thereby enriching the understanding of POC dynamics in this area. [ABSTRACT FROM AUTHOR]

Land-use and land-cover change (LULCC) is a critical environmental issue that has significant effects on biodiversity, ecosystem services, and climate change. This study examines the land-use and land-cover (LULC) spatiotemporal dynamics across a three-decade period (1998–2023) in a district area. In order to forecast the LULCC patterns, this study suggests a hybrid strategy that combines the random forest method with multi-layer perceptron (MLP) and Markov chain analysis. To predict the dynamics of LULC changes for the year 2035, a hybrid technique based on multi-layer perceptron and Markov chain model analysis (MLP-MCA) was employed. The area of developed land has increased significantly, while the amount of bare land, vegetation, and forest cover have all decreased. This is because the principal land types have changed due to population growth and economic expansion. This study also discovered that between 1998 and 2023, the built-up area increased by 468 km2 as a result of the replacement of natural resources. It is estimated that 25.04% of the study area's urbanization will increase by 2035. The performance of the model was confirmed with an overall accuracy of 90% and a kappa coefficient of around 0.89. It is important to use advanced predictive models to guide sustainable urban development strategies. The model provides valuable insights for policymakers, land managers, and researchers to support sustainable land-use planning, conservation efforts, and climate change mitigation strategies. [ABSTRACT FROM AUTHOR]

Prevalence estimates of Parkinson's disease (PD)—the fastest-growing neurodegenerative disease—are generally underestimated due to issues surrounding diagnostic accuracy, symptomatic undiagnosed cases, suboptimal prodromal monitoring, and limited screening access. Remotely monitored wearable devices and sensors provide precise, objective, and frequent measures of motor and non-motor symptoms. Here, we used consumer-grade wearable device and sensor data from the WATCH-PD study to develop a PD screening tool aimed at eliminating the gap between patient symptoms and diagnosis. Early-stage PD patients (n = 82) and age-matched comparison participants (n = 50) completed a multidomain assessment battery during a one-year longitudinal multicenter study. Using disease- and behavior-relevant feature engineering and multivariate machine learning modeling of early-stage PD status, we developed a highly accurate (92.3%), sensitive (90.0%), and specific (100%) random forest classification model (AUC = 0.92) that performed well across environmental and platform contexts. These findings provide robust support for further exploration of consumer-grade wearable devices and sensors for global population-wide PD screening and surveillance. [ABSTRACT FROM AUTHOR]

Building Performance Simulation extensively uses statistical learning techniques for quicker insights and improved accessibility. These techniques help understand the relationship between input variables and the desired outputs, and they can predict unknown observations. Prediction becomes more informative with uncertainty quantification, which involves computing prediction intervals. Conformal prediction has emerged over the past 25 years as a flexible and rigorous method for estimating uncertainty. This approach can be applied to any pre-trained model, creating statistically rigorous uncertainty sets or intervals for model predictions. This study uses data from simulated buildings to demonstrate the powerful applications of conformal prediction in Building Performance Simulation (BPS) and, consequently, to the broader energy sector. Results show that conformal prediction can be applied when any assumptions about input and output variables are made, enhancing understanding and facilitating informed decision-making in energy system design and operation. [ABSTRACT FROM AUTHOR]

Previous studies were mostly conducted based on sparse time series and different phenological images, which often ignored the dramatic changes in salinization evolution throughout the year. Based on Landsat and moderate-resolution-imaging spectroradiometer (MODIS) images from 2000 to 2020, this study applied the Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM) algorithm to obtain similar phenological images for the month of April for the past 20 years. Based on the random forest algorithm, the surface parameters of the salinization were optimized, and the feature space index models were constructed. Combined with the measured ground data, the optimal monitoring index model of salinization was determined, and then the spatiotemporal evolution patterns of salinization and its driving mechanisms in the Yellow River Delta were revealed. The main conclusions were as follows: (1) The derived long-time-series and similar phenological-fusion images enable us to reveal the patterns of change in the dramatic salinization in the year that we examined using the ESTARFM algorithm. (2) The NDSI-TGDVI feature space salinization monitoring index model based on point-to-point mode had the highest accuracy of 0.92. (3) From 2000 to 2020, the soil salinization in the Yellow River Delta showed an aggravating trend. The average value of salinization during the past 20 years was 0.65, which is categorized as severe salinization. The degree of salinization gradually decreased from the northeastern coastal area to the southwestern inland area. (4) The dominant factors affecting soil salinization in different historical periods varied. The research results could provide support for decision-making regarding the precise prevention and control of salinization in the Yellow River Delta. [ABSTRACT FROM AUTHOR]

Tree species classification using hyperspectral imagery shows incredible promise in developing a large-scale, high-resolution model for identifying tree species, providing unprecedented details on global tree species distribution. Many questions remain unanswered about the best practices for creating a global, general hyperspectral tree species classification model. This study aims to address three key issues in creating a hyperspectral species classification model. We assessed the effectiveness of three data-labeling methods to create training data, three data-splitting methods for training/validation/testing, and machine-learning and deep-learning (including semi-supervised deep-learning) models for tree species classification using hyperspectral imagery at National Ecological Observatory Network (NEON) Sites. Our analysis revealed that the existing data-labeling method using the field vegetation structure survey performed reasonably well. The random tree data-splitting technique was the most efficient method for both intra-site and inter-site classifications to overcome the impact of spatial autocorrelation to avoid the potential to create a locally overfit model. Deep learning consistently outperformed random forest classification; both semi-supervised and supervised deep-learning models displayed the most promising results in creating a general taxa-classification model. This work has demonstrated the possibility of developing tree-classification models that can identify tree species from outside their training area and that semi-supervised deep learning may potentially utilize the untapped terabytes of unlabeled forest imagery. [ABSTRACT FROM AUTHOR]

Soil nutrient content varies spatially across agricultural fields in hard-to-predict ways, particularly in floodplains with complex fluvial depositional history. Satellite reflectance data from the Sentinel-2 (S2) mission provides spatially continuous land reflectance data that can aid model development when used with point observations of nutrients. Reflectance from vegetation is assumed to obstruct land reflectance of bare soil, such that researchers have masked vegetation in models. We developed a routine for masking vegetation within Google Earth Engine (GEE) using Random Forest classification for iterative application to libraries of S2-images. Using gradient boosting, we then developed soil nutrient models for surface soils at a 250-ha agricultural site using S2 images. Soils were sampled at 2145 point locations to a 23-cm depth and analyzed for Ca, K, Mg, P, pH, S, and Zn. Results showed that masking vegetation improved model performance for models from subsets of the data (80% of samples used for model development, 20% validation), but full data sets did not require masking to achieve accuracy. Models of Ca, K, Mg, and S were successful (validation R2 > 0.60 to 0.96), but models for pH, P, and Zn failed. Bare soil composite images from S2 data are helpful in predicting soil fertility in low-relief floodplains. [ABSTRACT FROM AUTHOR]

Accurate prediction of lake surface water temperature (LSWT) is essential for understanding the impacts of climate change on aquatic ecosystems and for guiding environmental management strategies. Predictions of LSWT for two prominent lakes in northern China, Qinghai Lake and Hulun Lake, under various future climate scenarios, were conducted in the present study. Utilizing historical hydrometeorological data and MODIS satellite observations (MOD11A2), we employed three advanced machine learning models—Random Forest (RF), XGBoost, and Multilayer Perceptron Neural Network (MLPNN)—to predict monthly average LSWT across three future climate scenarios (ssp119, ssp245, ssp585) from CMIP6 projections. Through the comparison of training and validation results of the three models across both lake regions, the RF model demonstrated the highest accuracy, with a mean MAE of 0.348 °C and an RMSE of 0.611 °C, making it the most optimal and suitable model for this purpose. With this model, the predicted LSWT for both lakes reveals a significant warming trend in the future, particularly under the high-emission scenario (ssp585). The rate of increase is most pronounced under ssp585, with Hulun Lake showing a rise of 0.55 °C per decade (R2 = 0.72) and Qinghai Lake 0.32 °C per decade (R2 = 0.85), surpassing trends observed under ssp119 and ssp245. These results underscore the vulnerability of lake ecosystems to future climate change and provide essential insights for proactive climate adaptation and environmental management. [ABSTRACT FROM AUTHOR]

Accurate river ice mapping is crucial for predicting and managing floods caused by ice jams and for the safe operation of hydropower and water resource facilities. Although satellite multispectral images are widely used for river ice mapping, atmospheric contamination limits their effectiveness. This study developed river ice mapping models for the Han River in South Korea using atmospherically uncorrected Landsat-8 Operational Land Imager (OLI) multispectral reflectance data, addressing atmospheric influences with a Random Forest (RF) classification approach. The RF-based river ice mapping models were developed by implementing various combinations of input variables, incorporating the Landsat-8 multispectral top-of-atmosphere (TOA) reflectance, normalized difference indices for snow, water, and bare ice, and atmospheric factors such as aerosol optical depth, water vapor content, and ozone concentration from the Moderate Resolution Imaging Spectroradiometer observations, as well as surface elevation from the GLO-30 digital elevation model. The RF model developed using all variables achieved excellent performance in the classification of snow-covered ice, snow-free ice, and water, with an overall accuracy and kappa coefficient exceeding 98.4% and 0.98 for test samples, and higher than 83.7% and 0.75 when compared against reference river ice maps generated by manually interpreting the Landsat-8 images under various atmospheric conditions. The RF-based river ice mapping model for the atmospherically corrected Landsat-8 multispectral surface reflectance was also developed, but it showed very low performance under atmospheric conditions heavily contaminated by aerosol and water vapor. Aerosol optical depth and water vapor content were identified as the most important variables. This study demonstrates that multispectral reflectance data, despite atmospheric contamination, can be effectively used for river ice monitoring by applying machine learning with atmospheric auxiliary data to mitigate atmospheric effects. [ABSTRACT FROM AUTHOR]

Determining the responses of non-photosynthetic vegetation (NPV) and photosynthetic vegetation (PV) communities to climate change is crucial in illustrating the sensitivity and sustainability of these ecosystems. In this study, we evaluated the accuracy of inverting NPV and PV using Landsat imagery with random forest (RF), backpropagation neural network (BPNN), and fully connected neural network (FCNN) models. Additionally, we inverted MODIS NPV and PV time-series data using spectral unmixing. Based on this, we analyzed the responses of NPV and PV to precipitation and drought across different ecological regions. The main conclusions are as follows: (1) In NPV remote sensing inversion, the softmax activation function demonstrates greater advantages over the ReLU activation function. Specifically, the use of the softmax function results in an approximate increase of 0.35 in the R2 value. (2) Compared with a five-layer FCNN with 128 neurons and a three-layer BPNN with 12 neurons, a random forest model with over 50 trees and 5 leaf nodes provides better inversion results for NPV and PV (R2_RF-NPV = 0.843, R2_RF-PV = 0.861). (3) Long-term drought or heavy rainfall events can affect the utilization of precipitation by NPV and PV. There is a high correlation between extreme precipitation events following prolonged drought and an increase in PV coverage. (4) Under long-term drought conditions, the vegetation in the study area responded to precipitation during the last winter and growing season. This study provides an illustration of the response of semi-arid ecosystems to drought and wetting events, thereby offering a data basis for the effect evaluation of afforestation projects. [ABSTRACT FROM AUTHOR]

Global warming and climate change are significantly impacting local climates, causing more intense heat during the summer season, which poses risks to individuals with pre-existing health conditions and negatively affects overall human health. While various studies have examined the Surface Urban Heat Island (SUHI) phenomenon, these studies often focus on small to large geographic regions using low-to-moderate-resolution data, highlighting general thermal trends across large administrative areas. However, there is a growing need for methods that can detect microscale thermal patterns in environments familiar to urban residents, such as streets and alleys. The temperature-humidity index (THI), which incorporates both temperature and humidity data, serves as a critical measure of human-perceived heat. However, few studies have explored microscale THI variations within urban settings and identified potential THI hotspots at a local level where SUHI effects are pronounced. This research aims to address this gap by estimating THI at a finer resolution scale using data from multiple sensor platforms. We developed a model with the random forest algorithm to assess THI trends at a resolution of 0.5 m, utilizing various variables from different sources, including Landsat 8 land surface temperature (LST), unmanned aerial system (UAS)-derived LST, Sentinel-2 NDVI and NDMI, a wind exposure index, solar radiation modeled from aircraft and UAS-derived Digital Surface Models, and vehicle density and building floor area from social big data. Two models were constructed with different variables: Modelnatural, which includes variables related to only natural factors, and Modelmix, which includes all variables, including anthropogenic factors. The two models were compared to reveal how each source contributes to the model development and SUHI effects. The results show significant improvements, as Modelnatural had a fitting R2 = 0.5846, a root mean square error (RMSE) = 0.5936 and a mean absolute error (MAE) = 0.4294. Moreover, when anthropogenic factors were introduced, Modelmix performed even better, with R2 = 0.9638, RMSE = 0.1751, and MAE = 0.1065 (n = 923). This study contributes to the future of microscale SUHI analysis and offers important insights into urban planning and smart city development. [ABSTRACT FROM AUTHOR]

Detecting clouds, snow, and lakes in remote sensing images is vital due to their propensity to obscure underlying surface information and hinder data extraction. In this study, we utilize Sentinel-2 images to implement a two-stage random forest (RF) algorithm for image labeling and delve into the factors influencing neural network performance across six aspects: model architecture, encoder, learning rate adjustment strategy, loss function, input image size, and different band combinations. Our findings indicate the Feature Pyramid Network (FPN) achieved the highest MIoU of 87.14%. The multi-head self-attention mechanism was less effective compared to convolutional methods for feature extraction with small datasets. Incorporating residual connections into convolutional blocks notably enhanced performance. Additionally, employing false-color images (bands 12-3-2) yielded a 4.86% improvement in MIoU compared to true-color images (bands 4-3-2). Notably, variations in model architecture, encoder structure, and input band combination had a substantial impact on performance, with parameter variations resulting in MIoU differences exceeding 5%. These results provide a reference for high-precision segmentation of clouds, snow, and lakes and offer valuable insights for applying deep learning techniques to the high-precision extraction of information from remote sensing images, thereby advancing research in deep neural networks for semantic segmentation. [ABSTRACT FROM AUTHOR]

LiDAR-unmanned aerial system (LiDAR-UAS) technology can accurately and efficiently obtain detailed and accurate three-dimensional spatial information of objects. The classification of objects in estuarine areas is highly important for management, planning, and ecosystem protection. Owing to the presence of slopes in estuarine areas, distinguishing between dense vegetation (lawns and trees) on slopes and the ground at the tops of slopes is difficult. In addition, the imbalance in the number of point clouds also poses a challenge for accurate classification directly from point cloud data. A multifeature-assisted and multilayer fused neural network (MLF-PointNet++) is proposed for LiDAR-UAS point cloud classification in estuarine areas. First, the 3D shape features that characterize the geometric characteristics of targets and the visible-band difference vegetation index (VDVI) that can characterize vegetation distribution are used as auxiliary features to enhance the distinguishability of dense vegetation (lawns and trees) on slopes and the ground at the tops of slopes. Second, to enhance the extraction of target spatial information and contextual relationships, the feature vectors output by different layers of set abstraction in the PointNet++ model are fused to form a combined feature vector that integrates low and high-level information. Finally, the focal loss function is adopted as the loss function in the MLF-PointNet++ model to reduce the effect of imbalance in the number of point clouds in each category on the classification accuracy. A classification evaluation was conducted using LiDAR-UAS data from the Moshui River estuarine area in Qingdao, China. The experimental results revealed that MLF-PointNet++ had an overall accuracy (OA), mean intersection over union (mIOU), kappa coefficient, precision, recall, and F1-score of 0.976, 0.913, 0.960, 0.953, 0.953, and 0.953, respectively, for object classification in the three representative areas, which were better than the corresponding values for the classification methods of random forest, BP neural network, Naive Bayes, PointNet, PointNet++, and RandLA-Net. The study results provide effective methodological support for the classification of objects in estuarine areas and offer a scientific basis for the sustainable development of these areas. [ABSTRACT FROM AUTHOR]

Cytokine-induced neutrophil chemoattractant 1 (CINC-1), a cluster of differentiation 95 (CD95), fractalkine, and T-cell immunoglobulin and mucin domain 1 (TIM-1) are circulating proteins known to be involved in inflammation. While their roles have been studied in neurological conditions and cardiovascular diseases, their potential as peripheral artery disease (PAD) biomarkers remain unexplored. We conducted a cross-sectional diagnostic study using data from 476 recruited patients (164 without PAD and 312 with PAD). Plasma levels of CINC-1, CD95, fractalkine, and TIM-1 were measured at baseline. A PAD diagnosis was established at recruitment based on clinical exams and investigations, defined as an ankle-brachial index < 0.9 or toe-brachial index < 0.67 with absent/diminished pedal pulses. Using 10-fold cross-validation, we trained a random forest algorithm, incorporating clinical characteristics and biomarkers that showed differential expression in PAD versus non-PAD patients to predict a PAD diagnosis. Among the proteins tested, CINC-1, CD95, and fractalkine were elevated in PAD vs. non-PAD patients, forming a 3-biomarker panel. Our predictive model achieved an AUROC of 0.85 for a PAD diagnosis using clinical features and this 3-biomarker panel. By combining the clinical characteristics with these biomarkers, we developed an accurate predictive model for a PAD diagnosis. This algorithm can assist in PAD screening, risk stratification, and guiding clinical decisions regarding further vascular assessment, referrals, and medical/surgical management to potentially improve patient outcomes. [ABSTRACT FROM AUTHOR]

Approximately 15% of patients with colorectal cancer (CRC) exhibit a distinct molecular phenotype known as microsatellite instability (MSI). Accurate and non-invasive prediction of MSI status is crucial for cost savings and guiding clinical treatment strategies. The retrospective study enrolled 307 CRC patients between January 2020 and October 2022. Preoperative images of computed tomography and postoperative status of MSI information were available for analysis. The stratified fivefold cross-validation was used to avoid sample bias in grouping. Feature extraction and model construction were performed as follows: first, inter-/intra-correlation coefficients and the least absolute shrinkage and selection operator algorithm were used to identify the most predictive feature subset. Subsequently, multiple discriminant models were constructed to explore and optimize the combination of six feature preprocessors (Box-Cox, Yeo-Johnson, Max-Abs, Min–Max, Z-score, and Quantile) and three classifiers (logistic regression, support vector machine, and random forest). Selecting the one with the highest average value of the area under the curve (AUC) in the test set as the radiomics model, and the clinical screening model and combined model were also established using the same processing steps as the radiomics model. Finally, the performances of the three models were evaluated and analyzed using decision and correction curves.We observed that the logistic regression model based on the quantile preprocessor had the highest average AUC value in the discriminant models. Additionally, tumor location, the clinical of N stage, and hypertension were identified as independent clinical predictors of MSI status. In the test set, the clinical screening model demonstrated good predictive performance, with the average AUC of 0.762 (95% confidence interval, 0.635–0.890). Furthermore, the combined model showed excellent predictive performance (AUC, 0.958; accuracy, 0.899; sensitivity, 0.929) and favorable clinical applicability and correction effects. The logistic regression model based on the quantile preprocessor exhibited excellent performance and repeatability, which may further reduce the variability of input data and improve the model performance for predicting MSI status in CRC. [ABSTRACT FROM AUTHOR]

Hazel dormice Muscardinus avellanarius have severely declined since 2000 leading to increased legislative protection in the UK and Europe. Artificial nestboxes are widely used for its conservation and monitoring. Previous research has focused on how to identify suitable areas for nestboxes, but where to place individual boxes to promote occupancy is less well understood. Here, we demonstrate the use of machine learning Random Forest regression to predict nestbox occupancy from a wide range of microhabitat variables using a UK woodland as a case study. Random forest models are powerful predictive tools that allow simultaneous testing of many predictors with relatively few observations. Field data included observed nestbox occupancy (2017–2021) and measurements of 76 microhabitat variables collected in the summer of 2021 from 45 occupied and unused nestboxes located in a deciduous woodland in Berkshire, UK. We applied Random Forest regression to identify important variables and predict nestbox occupancy demonstrating robust approaches to tune model hyperparameters and evaluate importance metrics. In our study area, nestboxes were more likely to be occupied in sites with more hazel Corylus avellana, greater overall tree abundance but not fully closed canopies (optimal 80–85%), more honeysuckle Lolium periclymenum and hawthorn Crataegus monogyna, and when located further from footpaths and woodland margins. Occupancy over the study period was well predicted using microhabitat variables (13.3% OOB error) but future occupancy was more uncertain (33.3% error for 2021–2023 records). Modelling approaches that allow consideration of numerous variables from few locations or observations can be help identify relevant features and predict desirable outcomes of conservation actions. Here we demonstrate this approach identifying microhabitat variables that influence artificial nestbox occupancy by hazel dormice in a UK woodland. Findings offer some recommendations for local management that could promote nestbox occupancy and improve monitoring and conservation efforts. [ABSTRACT FROM AUTHOR]

This study develops an intelligent optimization method of the facility environment (i.e., road facilities and surrounding landscapes) from drivers' visual perception to adjust operation speeds on rural roads. Different from previous methods that heavily rely on expert experience and are time‐consuming, this method can rapidly generate optimized visual images of the facility environment and promptly verify the optimization effects. In this study, a visual road schema model is established to quantify the facility environment from drivers' visual perception, and an automated optimization scheme determination approach considering the original facility environment characteristics is proposed using self‐explaining theory. Then, Cycle‐consistent generative adversarial network is used to automatically generate optimized facility environment images. To verify the optimization effect, operation speeds of the optimized facility environments are predicted using random forest. The case study shows that this method can effectively optimize the facility environment where original operation speeds are more than 20% over the speed limits, and the whole process only takes 1 h far less than several months or years in previous ways. Overall, this study advances the intelligence level in optimizing the facility environment and enhances rural road safety. [ABSTRACT FROM AUTHOR]