A Robust Machine Learning Framework for Diabetes Prediction

Background and aimOver the past decades, we have witnessed an immense expansion in the arsenal and performance of machine learning (ML) algorithms. One of the most important fields that could benefit from these advancements is biomedical science. To streamline the training and evaluation of binary classifiers, we constructed a universal and flexible ML framework that uses tabular biomedical data as input.Methods and resultsOur framework requires the input data to be provided as a comma-separated values file, in which rows correspond to subjects and columns represent different features. After reading the content of this file, the framework enables the users to perform outlier detection, handle missing values, rescale features, and tackle class imbalance. Then, hyperparameter tuning, feature selection, and internal validation are performed using nested cross-validation. If an additional dataset is available, the framework also provides the option for external validation. Users may also compute SHapley Additive exPlanations values to interpret the individual predictions of the model and identify the most important features. Our ML framework was implemented in Python (version 3.9), and its source code is freely available via GitHub. In the second part of this paper, we also demonstrate the usage of the framework through a case study from the field of cardiovascular imaging.ConclusionsThe proposed ML framework enables the efficient training and evaluation of binary classifiers on tabular biomedical data. We hope our framework will serve as a useful resource for both learning and research purposes and will promote further innovation.

A tensile properties-related fatigue strength predicted machine learning framework for alloys used in aerospace

In this paper, the risk pattern of e-bike riders in China was examined, based on tree-structured machine learning techniques. Three-year crash/violation data were acquired from the Kunshan traffic police department, China. Firstly, high-risk (HR) electric bicycle (e-bike) riders were defined as those with at-fault crash involvement, while others (i.e., non-at-fault or without crash involvement) were considered as non-high-risk (NHR) riders, based on quasi-induced exposure theory. Then, for e-bike riders, their demographics and previous violation-related features were developed based on the crash/violation records. After that, a systematic machine learning (ML) framework was proposed so as to capture the complex risk patterns of those e-bike riders. An ensemble sampling method was selected to deal with the imbalanced datasets. Four tree-structured machine learning methods were compared, and a gradient boost decision tree (GBDT) appeared to be the best. The feature importance and partial dependence were further examined. Interesting findings include the following: (1) tree-structured ML models are able to capture complex risk patterns and interpret them properly; (2) spatial-temporal violation features were found as important indicators of high-risk e-bike riders; and (3) violation behavior features appeared to be more effective than violation punishment-related features, in terms of identifying high-risk e-bike riders. In general, the proposed ML framework is able to identify the complex crash risk pattern of e-bike riders. This paper provides useful insights for policy-makers and traffic practitioners regarding e-bike safety improvement in China.

Human cytochrome P450 (CYP450) enzymes play a crucial role in drug metabolism and pharmacokinetics. CYP450 inhibition can lead to toxicity, in particular when drugs are co-administered with other drugs and xenobiotics or in the case of polypharmacy. Predicting CYP450 inhibition is also important for rational drug discovery and development, and precision in drug repurposing. In this overarching context, digital transformation of drug discovery and development, for example, using machine and deep learning approaches, offers prospects for prediction of CYP450 inhibition through computational models. We report here the development of a majority-voting machine learning framework to classify inhibitors and noninhibitors for seven major human liver CYP450 isoforms (CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2C19, CYP2D6, and CYP3A4). For the machine learning models reported herein, we employed interaction fingerprints that were derived from molecular docking simulations, thus adding an additional layer of information for protein-ligand interactions. The proposed machine learning framework is based on the structure of the binding site of isoforms to produce predictions beyond previously reported approaches. Also, we carried out a comparative analysis so as to identify which representation of test compounds (molecular descriptors, molecular fingerprints, or protein-ligand interaction fingerprints) affects the predictive performance of the models. This work underlines the ways in which the structure of the enzyme catalytic site influences machine learning predictions and the need for robust frameworks toward better-informed predictions.

Today, the use of artificial intelligence in electron optics, as in many other fields, has begun to increase. In this scope, we present a machine learning framework to predict experimental cross-section data. Our framework includes 8 deep learning models and 13 different machine learning algorithms that learn the fundamental structure of the data. This article aims to develop a machine learning framework to accurately predict double-differential cross-section values. This approach combines multiple models such as convolutional neural networks, machine learning algorithms, and autoencoders to create a more robust prediction system. The data for training the models are obtained from experimental data for different atomic and molecular targets. We developed a methodology for learning tasks, mainly using rigorous prediction error limits. Prediction results show that the machine learning framework can predict the scattering angle and energy of scattering electrons with high accuracy, with an R-squared score of up to 99% and a mean squared error of <0.7. This performance result demonstrates that the proposed machine learning framework can be used to predict electron scattering events, which could be useful for applications such as medical physics.

The design and optimization of advanced semiconductor devices, such as multi gate MOSFETs, play a critical role in the miniaturization process of integrated circuits and the advancement of modern electronics. As device geometries shrink into the sub 10nm region, accurate modeling and calibration of transistor behavior become increasingly essential yet computationally challenging. This study presents a machine learning (ML) driven predictive modeling approach that uses TCAD extracted datasets to accurately predict the drain current (Id) of nanowire and nanoribbon FETs. Six ML models were trained and evaluated on TCAD data XGBoost, Random Forest, 1D Convolutional Neural Network (CNN), Multilayer Perceptron (MLP), Long Short-Term Memory (LSTM), and TabNet. Among all models, the ensemble methods XGBoost and Random Forest performed the best, with R² scores of 0.9926 and 0.9549 and RMSE values of 2.1 × 10⁻⁴ A and 2.0 × 10⁻⁴ A, respectively. These results were significantly better than the deep learning models LSTM (R² = 0.6778) and 1D CNN (R² = 0.6997). These results demonstrate that tree-based models can generalize more effectively for structured TCAD data, challenging the prevailing assumption that deep learning models universally perform better in device level prediction tasks. The proposed ML framework significantly reduces the need for iterative TCAD simulations, enabling rapid and accurate prediction of device behavior. As a novel contribution, a fully integrated web-based application has been developed to deploy the best performing models for real time Id prediction and visualization.

The 100% renewable energy targets from the policymakers for the future grids have drawn a significant amount of interest. The high renewable penetration made future large interconnected grids (LIGs) more volatile and harder to understand using historical observations. This led to the need to study future LIGs using a wide range of future-year operating conditions and contingencies. However, existing planning tools are not sufficient for the dynamic security assessment (DSA) of these future LIGs due to a lack of detailed modeling capabilities and computational limitations when processing a wide range of scenarios. This paper addresses these two challenges by proposing; 1) an efficient modeling framework that can generate large grid’s dynamic data with cascade behaviors for a wide range of scenarios. This data is generated by respecting the constraints of production cost models and their respective AC power flow dynamic simulation models at an hourly resolution; 2) an unsupervised machine learning(ML)-based approach for fast scanning the DSA data. The proposed approach uses feature engineering techniques and fast Fourier transforms to transform the time series signals into visually distinguishable frequency domain signals for DSA. The proposed simulation framework is used to generate 1.485 terabytes of dynamic simulation data for the 2028 WECC system containing 4455 scenarios. The proposed ML framework used the 2028 WECC system to demonstrate its effectiveness and speed in identifying critical scenarios.

Multiscale monitoring of industrial chemical process using wavelet-entropy aided machine learning approach

Parkinson's disease (PD) presents a growing global health challenge, with early detection being crucial for effective management and treatment. This study seeks to develop an innovative machine learning (ML) framework for the early detection of PD by integrating advanced techniques for data preprocessing, dimensionality reduction, feature selection, and ensemble classification, aiming to significantly improve detection accuracy and timeliness. The research employs a robust ML pipeline, beginning with data preprocessing using mean imputation, standardization, min-max scaling, and SMOTE (Synthetic Minority Over-sampling Technique) to handle imbalanced data. Dimensionality reduction is achieved through Principal Component Analysis (PCA), while feature selection is performed using SelectKBest coupled with the ANOVA F-test to identify the most relevant features. Four ensemble methods—Random Forest, Gradient Boosting, XGBoost, and Support Vector Machine (SVM)—are evaluated for classification. Among the classifiers tested, the Gradient Boosting model stands out with an impressive accuracy of 0.9487, demonstrating its superior performance in PD detection. Integrating multiple preprocessing, dimensionality reduction, and feature selection techniques proves essential in optimizing model performance, highlighting the importance of a multifaceted approach in handling complex datasets. This research introduces a comprehensive ML framework that combines multiple advanced techniques in a streamlined process, significantly improving the early detection of Parkinson's disease. Ensemble methods, combined with strategic feature selection and data balancing techniques, offer a novel approach that could be applied to other neurodegenerative disorders, expanding its potential impact beyond PD detection.

Languages morphological context varies by community. The linguistic analysis became more complex due to grammatical variations, cultural, traditional, slang, misspellings, and language variance. Many studies in sentimental analysis have focused on natural language processing and peoples opinions. Text language processing takes time, requires lots of storage space, and a fast computer to work in distributed networks. Many developers choose Hadoop and Map Reduce to process Big Data. This study developed a methodology that employs Apache Spark as a text classification processing engine since it is faster in cluster computing systems. African libraries and packages for language lemmatization and stemming are still lacking. The proposed approach was utilized to detect offensive Swahili texts in social networks. Swahili is the third most widely spoken language in Africa. Four different machine learning techniques were tested as benchmarks, with the multinomial logistic model proving to be the most effective. The evaluation measures show that the proposed machine learning framework is versatile and suitable for usage in centralized and distributed systems.

Nighttime high-power aircraft engine run-ups represent a critical operational activity with potential implications for airport management and environmental assessment; yet, their quantitative prediction remains limited in existing literature. The objective of this study is to develop and evaluate a data-driven machine learning framework capable of accurately predicting nighttime high-power engine run-up characteristics using historical operational data. A supervised learning methodology is adopted using an open-access dataset comprising approximately 860 nighttime run-up records. The proposed framework includes data preprocessing, temporal feature engineering (month, season, and quarter), exploratory data analysis, correlation assessment, and comparative model development using Linear Regression, Random Forest, Gradient Boosting, Extreme Gradient Boosting (XGBoost), and an Artificial Neural Network (multilayer perceptron). Model performance is evaluated using coefficient of determination (R²), root mean square error (RMSE), mean absolute error, and mean absolute percentage error, supported by predicted-versus-actual comparisons and learning-curve analysis. The results demonstrate that ensemble-based models outperform linear and neural network approaches, with XGBoost achieving the highest predictive accuracy on the testing dataset (R² ≈ 0.90, RMSE ≈ 2.15). Learning curves further indicate stable validation performance with increasing training data size. The proposed machine learning framework provides a structured and data-driven approach for predicting nighttime high-power aircraft engine run-up activity within the analyzed operational dataset, offering a valuable analytical tool for data-driven assessment of airport ground operations.

Background/Objectives: Acute respiratory distress syndrome (ARDS) represents a critical complication in polytrauma patients, characterized by diffuse lung inflammation and bilateral pulmonary infiltrates with mortality rates reaching 45% in intensive care units (ICU). The heterogeneous nature of ARDS and complex clinical presentation in severely injured patients poses substantial diagnostic challenges, necessitating early prediction tools to guide timely interventions. Machine learning (ML) algorithms have emerged as promising approaches for clinical decision support, demonstrating superior performance compared to traditional scoring systems in capturing complex patterns within high-dimensional medical data. Based on the identified research gaps in early ARDS prediction for polytrauma populations, our study aimed to: (i) develop a balanced random forest (BRF) ML model for early ARDS prediction in critically ill polytrauma patients, (ii) identify the most predictive clinical features using ANOVA-based feature selection, and (iii) evaluate model performance using comprehensive metrics addressing class imbalance challenges. Methods: This retrospective cohort study analyzed 407 polytrauma patients admitted to the ICU of the Center of Traumatology and Major Burns of Ben Arous, Tunisia, between 2017 and 2021. We implemented a comprehensive ML pipeline that incorporates Tomek Links undersampling, ANOVA F-test feature selection for the top 10 predictive variables, and SMOTE oversampling with a conservative sampling rate of 0.3. The BRF classifier was trained with class weighting and evaluated using stratified 5-fold cross-validation. Performance metrics included AUROC, PR-AUC, sensitivity, specificity, F1-score, and Matthews correlation coefficient. Results: Among 407 patients, 43 developed ARDS according to the Berlin definition, representing a 10.57% incidence. The BRF model demonstrated exceptional predictive performance with an AUROC of 0.98, a sensitivity of 0.91, a specificity of 0.80, an F1-score of 0.84, and an MCC of 0.70. Precision-recall AUC reached 0.86, demonstrating robust performance despite class imbalance. During stratified cross-validation, AUROC values ranged from 0.93 to 0.99 across folds, indicating consistent model stability. The top 10 selected features included procalcitonin, PaO2 at ICU admission, 24-h pH, massive transfusion, total fluid resuscitation, presence of pneumothorax, alveolar hemorrhage, pulmonary contusion, hemothorax, and flail chest injury. Conclusions: Our BRF model provides a robust, clinically applicable tool for early prediction of ARDS in polytrauma patients using readily available clinical parameters. The comprehensive two-step resampling approach, combined with ANOVA-based feature selection, successfully addressed class imbalance while maintaining high predictive accuracy. These findings support integrating ML approaches into critical care decision-making to improve patient outcomes and resource allocation. External validation in diverse populations remains essential for confirming generalizability and clinical implementation.

Introduction: Heart failure (HF) in critical care settings exhibits complex interactions with glycemic dysregulation. Traditional metrics fail to capture dynamic glucose fluctuations linked to poor outcomes. This study defines ICU-specific glycemic targets and establishes a machine learning (ML) framework integrating novel glucose indicators to predict 28-day mortality. Methods: Cohort: 18,211 ICU HF patients (MIMIC-IV, 2008-2019). Analysis: 1. Glycemic Metrics: Calculated time-weighted average glucose (TWABG), time-in-range (TIR), time-below-range (TBR), time-above-range (TAR) 2. Association Studies: Restricted cubic splines (RCS) to identify optimal glycemic ranges. Multivariable Cox regression with adjustment and subgroup analysis with interaction testing. 3. ML Modeling: Dual-stage feature selection: Lasso regression and Boruta algorithm. Model development: Developed models using 14 algorithms; SHAP interpretation applied to best-performing model (highest ROC-AUC). Validation: SHAP analysis of an exemplar case demonstrated individualized mortality risk interpretation without formal model validation. Results: 1. Optimal Glycemic Ranges: Patients within the optimal TWABG range (85–133 mg/dL) demonstrated the lowest mortality (15.09%,P< 0.001), with improved survival associated with both higher time-in-range (TIR) proportion and absence of glycemic excursions (time-below-range [TBR] or time-above-range [TAR] exposure) 2. ML Prediction: Feature selection via Lasso regression identified 30 predictors. Boruta algorithm confirmed 29 significant features. The optimized XGBoost model predicted 28-day mortality with AUC 0.779 (0.762-0.796) Conclusions: Maintaining blood glucose within TWABG 85–133 mg/dL and maximizing TIR significantly reduces 28-day mortality in ICU HF patients, while avoiding excursions independently improves survival. Our ML framework, integrating dynamic glucose metrics with clinically selected features, accurately predicted mortality risk (AUC 0.779). Findings advocate ICU-specific glucose management and demonstrate ML’s potential for risk stratification and personalized interventions.

Classifying a particular affective state from the patterns of fMRI data is challenging. This is because of difficulty in finding discriminative and computationally inexpensive features and methods. In the present work, we employed a kernel-based machine learning framework utilizing covariance features and support vector machine (SVM) to classify 3 emotional states: pleasant (Ple), neutral (Neu), and unpleasant (Unp). We examined task functional magnetic resonance imaging (fMRI) of 5 healthy subjects who passively viewed a series of relevant pictures. Kernels (covariance matrix) per subject were extracted, mean-centered, and normalized by standard deviation. A standard leave-one-out cross validation was employed for generalization error. A balanced accuracy was computed to report overall accuracy. Discrimination maps derived from SVM weights were used to evaluate contributions of the input voxels. Best accuracy was achieved in Ple-vs-Neu with 88%, followed by Ple-vs-Unp (79%), and and Unp-vs-Neu (68%). Discrimination and t-maps suggested highly contributed regions close to dorsal attention network. The proposed machine learning framework is efficient and can be generalized to differentiate other cognitive tasks and emotions in fMRI.

Machine learning framework for predicting failure mode and shear capacity of ultra high performance concrete beams