Biochar-based persulfate activation: Rate constant prediction, key variables identification, and system optimization

Although the goal of rectal cancer treatment is to restore gastrointestinal continuity, some patients with rectal cancer develop a permanent stoma (PS) after sphincter-saving operations. Although many studies have identified the risk factors and causes of PS, few have precisely predicted the probability of PS formation before surgery. To validate whether an artificial intelligence model can accurately predict PS formation in patients with rectal cancer after sphincter-saving operations. Patients with rectal cancer who underwent a sphincter-saving operation at Taipei Medical University Hospital between January 1, 2012, and December 31, 2021, were retrospectively included in this study. A machine learning technique was used to predict whether a PS would form after a sphincter-saving operation. We included 19 routinely available preoperative variables in the artificial intelligence analysis. To evaluate the efficiency of the model, 6 performance metrics were utilized: accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiving operating characteristic curve. In our classification pipeline, the data were randomly divided into a training set (80% of the data) and a validation set (20% of the data). The artificial intelligence models were trained using the training dataset, and their performance was evaluated using the validation dataset. Synthetic minority oversampling was used to solve the data imbalance. A total of 428 patients were included, and the PS rate was 13.6% (58/428) in the training set. The logistic regression (LR), Gaussian Naïve Bayes (GNB), Extreme Gradient Boosting (XGB), Gradient Boosting (GB), random forest, decision tree and light gradient boosting machine (LightGBM) algorithms were employed. The accuracies of the logistic regression (LR), Gaussian Naïve Bayes (GNB), Extreme Gradient Boosting (XGB), Gradient Boosting (GB), random forest (RF), decision tree (DT) and light gradient boosting machine (LightGBM) models were 70%, 76%, 89%, 93%, 95%, 79% and 93%, respectively. The area under the receiving operating characteristic curve values were 0.79 for the LR model, 0.84 for the GNB, 0.95 for the XGB, 0.95 for the GB, 0.99 for the RF model, 0.79 for the DT model and 0.98 for the LightGBM model. The key predictors that were identified were the distance of the lesion from the anal verge, clinical N stage, age, sex, American Society of Anesthesiologists score, and preoperative albumin and carcinoembryonic antigen levels. Integration of artificial intelligence with available preoperative data can potentially predict stoma outcomes after sphincter-saving operations. Our model exhibited excellent predictive ability and can improve the process of obtaining informed consent.

The compressive strength (CS) of the hollow concrete masonry prism is known as an important parameter for designing masonry structures. In general, the CS is determined using laboratory tests, however, laboratory tests are time-consuming and high-cost. Thus, it is necessary to evaluate and estimate the CS using different methods, for example, machine learning techniques. This study employed Gradient Boosting (GB) to evaluate and predict the CS of hollow masonry prism. The database consists of 102 hollow concrete specimens taken from different previous published literature used for modeling. The output is the CS of the hollow masonry prism, while the inputs include the compressive strength of mortar (fm), the compressive strength of blocks (fb), height-to-thickness ratio (h/t), the ratio of fm/fb. To reduce the overfitting problem, this study used K-Fold cross-validation, then particle swarm optimization (PSO) was employed to obtain the optimum hyperparameter. The GB model then was modeled using the optimum hyperparameters. The results showed that the GB model performed very well in evaluating and predicting the CS of the hollow masonry prims with a high prediction accuracy, the values of R2, RMSE, MAE, and MAPE are 0.977, 0.803 MPa, 0.612 MPa, and 0.036%, respectively. The performance of the GB model in this study outperformed in comparison to six different machine learning models (decision tree, linear regression, random forest regression, ridge regression, Artificial Neural network, and Extreme Gradient Boosting) used in previous studies. The results of sensitivity analysis using SHAP and PDP-2D indicate that the CS is strongly dependent on the fb (with a mean SHAP value of 3.2), h/t (with a mean SHAP value of 1.63), while the fm/fb (with a mean SHAP value of 0.57) had a small effect on the CS. Thus, it can be stated that this research provides a good method to evaluate and predict the CS of the hollow masonry prism, which can bring good knowledge for practical application in this field.

The goal of this work is to show how machine learning models, such as the random forest, neural network, gradient boosting, and AdaBoost models, can be used to forecast the fatigue life (N) of plain concrete under uniaxial compression. Here, we developed our final machine learning model by generating the following three data files from the original data used in the work of Zhang et al.: (a) grouped data with the same input variable value and different output variable logN value, (b) data excluding outliers selected by three or more outlier detection methods; (c) average data excluding outliers, created by averaging the grouped data after excluding outliers from among the grouped data. Excluding the sustained strength of the concrete variable, originally treated as the seventh input variable in the work of Zhang et al., resulted in improving the determination coefficient (R2) values. Moreover, the gradient boosting model showed a high R2 value at 0.753, indicating a high accuracy in predicting outcomes. Further analysis using data excluding outliers shows that the R2 value increased to 0.803. Moreover, the average data excluding outliers provided the best R2 value at 0.915. Finally, a permutation feature importance (PFI) analysis was carried out to determine the strength of the relationship between the feature and the target value for the gradient boosting model. The analysis results showed that the maximum stress level (Smax) and loading frequency (f) were the most significant input variables, followed by compressive strength (f′c) and maximum to minimum stress ratio (R). Shape and height to width ratio (h/w) were the features with a non-significant influence on the model. This trend was previously confirmed by a Pearson and Spearman correlation analysis.

Overweight/obesity has become a critical global health issue, as these conditions are strongly associated with elevated risk of diabetes, stroke, cardiovascular disorders, and certain types of cancer. In recent decades, Bangladesh has faced a notable rise in overweight/obesity prevalence-women are more prone to obesity than men. This study presents a comprehensive strategy for identifying risk factors and predicting overweight and obesity through machine learning (ML) classifiers among ever-married Bangladeshi women aged 15-49 years. Data from the 2017-2018 BDHS, a nationally representative survey, were examined. The data were pre-processed and subsequently balanced using the synthetic minority over-sampling technique and edited nearest neighbors (SMOTE-ENN) approach. Various feature identification techniques, including Chi-Square, LASSO, and Sequential Forward Selection, were employed to determine the key risk features. Later, permutation feature importance and SHAP analysis were employed to assess the influence of these risk factors on overweight/obesity. The classification of overweight and obesity was conducted using seven machine learning models: Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), K-nearest Neighbors (KNN), eXtreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost), and Multilayer Perceptron (MLP). Among the evaluated models, SVM performed best, reaching 95.79% accuracy and 97.32% precision when combined with SMOTE-ENN and hyper-parameter tuning. The study found that key factors contributing to being overweight/obese include age, division, type of residence, educational levels of both the respondent and her partner, number of children, frequency of television viewing, and wealth status; where wealth status, age, and frequency of watching television have strong influences. Therefore, integrating the balancing algorithm with the embedded feature selection strategy was effective in classifying overweight/obese women and could enhance decision-making for preventive measures in public health through timely predictions of overweight/obesity.

Overweight/obesity has become a critical global health issue, as these conditions are strongly associated with elevated risk of diabetes, stroke, cardiovascular disorders, and certain types of cancer. In recent decades, Bangladesh has faced a notable rise in overweight/obesity prevalence—women are more prone to obesity than men. This study presents a comprehensive strategy for identifying risk factors and predicting overweight and obesity through machine learning (ML) classifiers among ever-married Bangladeshi women aged 15–49 years. Data from the 2017–2018 BDHS, a nationally representative survey, were examined. The data were pre-processed and subsequently balanced using the synthetic minority over-sampling technique and edited nearest neighbors (SMOTE-ENN) approach. Various feature identification techniques, including Chi-Square, LASSO, and Sequential Forward Selection, were employed to determine the key risk features. Later, permutation feature importance and SHAP analysis were employed to assess the influence of these risk factors on overweight/obesity. The classification of overweight and obesity was conducted using seven machine learning models: Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), K-nearest Neighbors (KNN), eXtreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost), and Multilayer Perceptron (MLP). Among the evaluated models, SVM performed best, reaching 95.79% accuracy and 97.32% precision when combined with SMOTE-ENN and hyper-parameter tuning. The study found that key factors contributing to being overweight/obese include age, division, type of residence, educational levels of both the respondent and her partner, number of children, frequency of television viewing, and wealth status; where wealth status, age, and frequency of watching television have strong influences. Therefore, integrating the balancing algorithm with the embedded feature selection strategy was effective in classifying overweight/obese women and could enhance decision-making for preventive measures in public health through timely predictions of overweight/obesity.

Rice is one of the six priority crops chosen by the government of Rwanda to increase agricultural productivity. This crop faces significant challenges from climate variability and soil conditions, affecting yields and food security. The past studies concentrated on building machine learning regression models for predicting the rice productivity with the help of weather parameters as predictors. Few researchers incorporated some soil nutrients to improve the precision of the prediction model. However, most of the farmers in Rwanda rely on agricultural input like inorganic fertilizer to boost the agriculture productivity. This means that the weather parameters are not sufficient to build a robust crop yield prediction model without incorporating the soil parameters. In this paper, rice yield status (favorable, unfavorable) prediction model was optimized using machine learning techniques, including logistic regression, random forest, gradient boosting, extreme gradient boosting, support vector machine, and artificial neural networks. The models were trained on historical weather parameters and soil nutrient data collected from the National Institute of Statistics of Rwanda, Rwanda Agriculture and Animal Resources Development Board and Rwanda Meteorology Agency. Findings indicated that the eXtreme Gradient Boosting (XGBOOST) model outperformed other models under consideration with 91%, 94.1%, and 82.7% of prediction accuracy, sensitivity, and precision, respectively. The above results demonstrate that the XGBOOST is the most robust model in predicting rice yield. The feature importance technique applied in this study also revealed that weather parameters (rainfall, and temperature), phosphorus, sulfur, manganese and carbon were crucial factors to prioritize for rice yield prediction compared to other model predictors under study. The results may provide valuable insights into whether the rice yield will be classified as low or high. This information is essential for helping farmers and decision-makers obtain actionable insights into expected crop yields, enabling them to make informed decisions on agricultural inputs resource allocation to maximize Rwanda's agricultural productivity. Key words: artificial intelligence, machine learning, precision farming, Smart agriculture, yield prediction

<div class="section abstract"><div class="htmlview paragraph">Light-duty vehicle emissions regulations worldwide impose stringent limits on particulate matter (PM) emissions, necessitating accurate modelling and prediction of particulate emissions across a range of sizes (as low as 10 nm). It has been shown that the decision tree-based ensemble machine learning technique known as Random Forest can accurately predict particle size, concentration, and accumulation mode geometric standard deviation (GSD) for particulate emission diameters as low as 23 nm from a highly boosted gasoline direct injection (GDI) engine operating on a single fuel, while also offering insights into the underlying factors of emissions production because of the interpretable nature of decision trees.</div><div class="htmlview paragraph">This work builds on the prior Random Forest research as its basis and further investigates the relative performance of five decision tree-based machine learning techniques in predicting these particulate emission parameters and extends the work to 10 nm particles. In addition to Random Forest, the selected techniques consist of four gradient boosting models: GBM, XGBoost, LightGBM, and CatBoost. Moreover, the influences of fuel chemistry are assessed by using data from 13 gasoline fuel blends, including blends with ethanol and methanol – common bio- and e-fuels. The results show that the CatBoost model achieves the highest prediction accuracy (R<sup>2</sup> between 0.77 and 0.932), even when the feature set is reduced to improve computational efficiency. Random Forest and LightGBM are also shown to be suitable for PM emissions estimation. Permutation feature importance was used to highlight the dependence of PM emissions on both fuel and engine operating parameters – offering new insights into the effect of fuel properties on particulate emissions and their formation in highly boosted engines.</div></div>

ObjectivesTo evaluate the efficacy of the Cox-Maze IV procedure (CMP-IV) in combination with valve surgery in patients with both atrial fibrillation (AF) and valvular disease and use machine learning algorithms to identify potential risk factors of AF recurrence.MethodsA total of 1,026 patients with AF and valvular disease from two hospitals were included in the study. 555 patients received the CMP-IV procedure in addition to valve surgery and left atrial appendage ligation (CMP-IV group), while 471 patients only received valve surgery and left atrial appendage ligation (Non-CMP-IV group). Kaplan–Meier analysis was used to calculate the sinus rhythm maintenance rate. 58 variables were selected as variables for each group and 10 machine learning models were developed respectively. The performance of the models was evaluated using five-fold cross-validation and metrics including F1 score, accuracy, precision, and recall. The four best-performing models for each group were selected for further analysis, including feature importance evaluation and SHAP analysis.ResultsThe 5-year sinus rhythm maintenance rate in the CMP-IV group was 82.13% (95% CI: 78.51%, 85.93%), while in the Non-CMP-IV group, it was 13.40% (95% CI: 10.44%, 17.20%). The eXtreme Gradient Boosting (XGBoost), LightGBM, Category Boosting (CatBoost) and Random Fores (RF) models performed the best in the CMP-IV group, with area under the curve (AUC) values of 0.768 (95% CI: 0.742, 0.786), 0.766 (95% CI: 0.744, 0.792), 0.762 (95% CI: 0.723, 0.801), and 0.732 (95% CI: 0.701, 0.763), respectively. In the Non-CMP-IV group, the LightGBM, XGBoost, CatBoost and RF models performed the best, with AUC values of 0.738 (95% CI: 0.699, 0.777), 0.732 (95% CI: 0.694, 0.770), 0.724 (95% CI: 0.668, 0.789), and 0.716 (95% CI: 0.656, 0.774), respectively. Analysis of feature importance and SHAP revealed that duration of AF, preoperative left ventricular ejection fraction, postoperative heart rhythm, preoperative neutrophil-lymphocyte ratio, preoperative left atrial diameter and heart rate were significant factors in AF recurrence.ConclusionCMP-IV is effective in treating AF and multiple machine learning models were successfully developed, and several risk factors were identified for AF recurrence, which may aid clinical decision-making and optimize the individual surgical management of AF.

Breast cancer is the most incident cancer worldwide. The evaluation of molecular subtypes and their biomarkers plays an essential role in prognosis. The biomarkers used are Estrogen Receptor (ER), Progesterone Receptor (PR), Human Epidermal growth factor Receptor-type 2 (HER2), and Ki67. Based on these, subtypes are classified as Luminal A (LA), Luminal B (LB), HER2 subtype, and Triple-Negative Breast Cancer (TNBC). The gold standard for this analysis is histology and immunohistochemistry, semi-quantitative techniques that present inter-laboratory and inter-observer variations. The Fourier Transform Infrared micro-spectroscopy (micro-FTIR), which provides hyperspectral images with biochemical information of biological tissues, is applied together with artificial intelligence (AI) for cancer evaluation. In this thesis, twenty samples of two breast cancer cell lines, BT-474 and SK-BR-3, were used to define the optimal number of co-added scans for machine learning (ML) techniques. Linear Discriminant Analysis (LDA), Partial Least Squares Discriminant Analysis (PLS-DA), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGB) models were used. Sixty hyperspectral images of 320x320 pixels were collected from thirty patients of a human breast biopsies microarray, each containing a breast cancer (CA) and an adjacent tissue (AT) core. Automated methods based on K-Means clustering were developed for data organization and pre-processing to one-dimensional (1D) and two-dimensional (2D) data. The dataset was used to train two new deep learning models for breast cancer subtype evaluation: CaReNet-V1, a 1D Convolutional Neural Network (CNN); and CaReNet-V2, a 2D CNN. All ML models achieved similar performances with the b256_064 (256 background scans and 64 sample scans), b256_128, and b128_128 groups, where the best accuracy of 0.995 was presented by the XGB model. The b256_064 was established as the ideal among the three due to the shortest acquisition time. The K-Means-based method enabled fully automated preprocessing and organization, improving data quality and optimizing CNN training. CaReNet-V1 effectively classified CA and AT (individual spectra test accuracy of 0.89), as well as HER2 and TNBC subtypes (0.83 and 0.86), with greater difficulty for LA and LB (0.74 and 0.68). The model enabled the evaluation of the most contributing wavenumbers to the predictions, providing a direct relationship with the biochemical content of the samples. CaReNet-V2 demonstrated better performance than 1D, with test accuracies above 0.84, and enabled the prediction of ER, PR, and HER2 levels, where borderline values showed lower performance (minimum accuracy of 0.54). The Ki67 percentage regression demonstrated an absolute mean error of 3.6%. On the other hand, its impact evaluation by wavenumber was inferior to 1D. Thus, this study indicates image-based AI techniques using micro-FTIR as potential providers of additional information to pathological reports, also serving as patient biopsy screening techniques.

Objective To construct a prediction model for post-anesthesia recovery after endoscopy and validate it. Methods Selection of 2400 surgical patients who underwent painless gastroenteroscopy at Sichuan Electric Power Hospital from September 2019 to September 2024. After data preprocessing, the Logistic regression, random forest (RF), support vector machine (SVM), decision tree (DT), and extreme gradient boosting (XGBoost) model methods were used to construct prediction models. The predictive accuracy of the machine learning models was evaluated by calculating the area under the receiver operating characteristic (ROC) curve, and the contribution of each feature to the prediction result was evaluated by using the Shapley additive explanations (SHAP) method. Results According to the performance indicators, the best model among the five models was the XGBoost regression model, and the nine indicators were found to be closely related to the recovery time of general anesthesia during the feature importance analysis. Conclusion The XGBoost model constructed using multiple data can analyze the complex relationships between variables and outcomes and has good predictive performance and clinical value.

Enhancing high-entropy alloy performance: Predictive modelling of wear rates with machine learning

Evaluating compressive strength of concrete made with recycled concrete aggregates using machine learning approach

Characterization and optimization of total solids content of drilling fluids is critical for the efficiency and success of drilling operations. Traditional solids content analysis methods, such as retort analysis, require substantial human intervention and time, which can lead to inaccuracies, time-management issues, and increased operational risks. To address these issues and complement automated rheological property measurements during drilling, this study aims to develop and validate a machine learning-based framework for estimating solids content in drilling fluids from readily available rheological parameters. A comprehensive data set was compiled from more than 1600 laboratory reports of drilling fluid analyses across 130 oil wells globally. Text mining packages were used to convert laboratory reports into a numeric data set which consists of drilling fluid properties. Pre-processing steps were taken to clean, filter and reduce the dimensionality of the data. Unrealistic measurements, and missing data were filtered out and several feature-selection approaches such as least absolute shrinkage and selection operator (LASSO) regression, permutation feature importance, and correlation coefficient matrix were employed to remove redundant variables from the data set. Cleaned and dimensionality-optimized data set was then used for machine-learning model development. Due to their ability to capture non-linear relationships among multiple variables, artificial neural networks employing the resilient backpropagation algorithm were selected to fit regression models for estimating total solids content. Various configurations were tested, including different numbers of layers and neurons, as well as alternative data partitioning schemes. Considering reasonable values of rheological properties based on domain knowledge, filtering was applied to remove outliers which indicate experimental error. As a result of combining the observations from different feature importance analyses and considering physical relationships between rheological variables, redundant features as well as reports with many missing values were removed from the data set. Five features, including mud weight, chloride content, plastic viscosity, oil-water ratio and API fluid loss, were selected to carry over into the modeling stage. As a result of experimenting with 120 different neural network configurations, a two-hidden layer neural network with 40-30 neurons in each hidden layer was selected. The model achieved an R2 value of 0.96 for the training set and 0.89 for the testing set. The root-mean-square error (RMSE) was 0.87% for training and 1.27% for testing, both of which were considered acceptable levels of error. An analysis of the network weights indicated that mud weight (MW) and oil–water ratio (OWR) were the most influential features, contributing more significantly to the model’s predictions compared to other input variables.

Blood pressure abnormality detection and Interpretation utilizing Explainable Artificial Intelligence

Feature selection plays a critical role in the performance of machine learning (ML)-based estimation techniques. Although feature selection has received moderate attention for tracking the state of charge (SOC) in batteries used for electric vehicles (EVs), it remains under-explored in the context of photovoltaic battery energy storage system (PV-BESS) applications. This study evaluates the significance of input features for SOC estimation in PV-BESS. We employ the permutation feature importance (PFI) method to assess the contribution of input features. The estimator model is implemented using a one-dimensional convolutional neural network (1D-CNN). The results show that although conventional features such as voltage, current, and temperature remain dominant predictors during both charging and discharging cycles, environmental features such as solar irradiance and humidity can also make a moderate contribution. Following the feature importance analysis, the developed 1D-CNN estimator is also employed for SOC tracking using the full feature set. The results demonstrate that the estimator achieves good performance, yielding low root mean squared errors (RMSEs).