Discriminating insulin resistance in middle-aged nondiabetic women using machine learning approaches DOI Creative Commons
Zailing Xing, Henian Chen, Amy C. Alman

et al.

AIMS Public Health, Journal Year: 2024, Volume and Issue: 11(2), P. 667 - 687

Published: Jan. 1, 2024

<abstract><sec> <title>Objective</title> <p>We employed machine learning algorithms to discriminate insulin resistance (IR) in middle-aged nondiabetic women.</p> </sec><sec> <title>Methods</title> <p>The data was from the National Health and Nutrition Examination Survey (2007–2018). The study subjects were 2084 women aged 45–64. analysis included 48 predictors. We randomly divided into training (n = 1667) testing 417) datasets. Four techniques IR: extreme gradient boosting (XGBoosting), random forest (RF), (GBM), decision tree (DT). area under curve (AUC) of receiver operating characteristic (ROC), accuracy, sensitivity, specificity, positive predictive value, negative F1 score compared as performance metrics select optimal technique.</p> <title>Results</title> XGBoosting algorithm achieved a relatively high AUC 0.93 dataset 0.86 IR using predictors followed by RF, GBM, DT models. After selecting top five build models, XGBoost with 0.90 (training dataset) (testing remained prediction model. SHapley Additive exPlanations (SHAP) values revealed associations between IR, namely BMI (strongly impact on IR), fasting glucose positive), HDL-C (medium negative), triglycerides glycohemoglobin positive). threshold for identifying 29 kg/m<sup>2</sup>, 100 mg/dL, 54.5 89 5.6% BMI, glucose, HDL-C, triglycerides, glycohemoglobin, respectively.</p> <title>Conclusion</title> demonstrated superior discriminating women, predictors.</p> </sec></abstract>

Language: Английский

Data-Intensive Inventory Forecasting with Artificial Intelligence Models for Cross-Border E-Commerce Service Automation DOI Creative Commons
Yuk Ming Tang, Ka Yin Chau, Yui‐yip Lau

et al.

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(5), P. 3051 - 3051

Published: Feb. 27, 2023

Building an adaptative, flexible, resilient, and reliable inventory management system provides a supply of cross-border e-commerce commodities, enhances chain members with flow products, fulfills ever-changing customer requirements, enables service automation. This study uses company as case to collect intensive data. The key process the AI approach for data forecasting framework is constructed. shows that model’s optimization needs be combined problems specific companies information analysis optimization. suggestions highlights processes AI-predicting model. XGBoost method demonstrates best performance in terms accuracy (RMSE = 46.64%) reasonable computation time (9 min 13 s). research can generalized used useful basis further implementing algorithms other enterprises. In doing so, this current trend logistics 4.0 solutions via adoption robust data-intensive artificial intelligence models As expected, findings improve alleviation bullwhip impact sustainable development. E-commerce enterprises may provide better plan their so minimize excess or stock-outs, sales strategies promotional marketing activities.

Language: Английский

Citations

39
A comparative analysis of boosting algorithms for chronic liver disease prediction DOI Creative Commons
Shahid Mohammad Ganie, Pijush Kanti Dutta Pramanik

Healthcare Analytics, Journal Year: 2024, Volume and Issue: 5, P. 100313 - 100313

Published: Feb. 23, 2024

Chronic liver disease (CLD) is a major health concern for millions of people all over the globe. Early prediction and identification are critical taking appropriate action at earliest stages disease. Implementing machine learning methods in predicting CLD can greatly improve medical outcomes, reduce burden condition, promote proactive preventive healthcare practices those risk. However, traditional has some limitations which be mitigated through ensemble learning. Boosting most advantageous approach. This study aims to performance available boosting techniques prediction. Seven popular algorithms Gradient (GB), AdaBoost, LogitBoost, SGBoost, XGBoost, LightGBM, CatBoost, two publicly datasets (Liver patient dataset (LDPD) Indian (ILPD)) dissimilar size demography considered this study. The features ascertained by exploratory data analysis. Additionally, hyperparameter tuning, normalisation, upsampling used predictive analytics. proportional importance every feature contributing algorithm assessed. Each algorithm's on both assessed using k-fold cross-validation, twelve metrics, runtime. Among five algorithms, GB emerged as best overall performer datasets. It attained 98.80% 98.29% accuracy rates LDPD ILPD, respectively. also outperformed other regarding metrics except

Language: Английский

Citations

10
Green synthesis, characterization, and hepatoprotective effect of zinc oxide nanoparticles from Moringa oleifera leaves in CCl4-treated albino rats DOI Creative Commons
Hossam S. El‐Beltagi,

Marwa Rageb,

Mahmoud M. El-Saber

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(9), P. e30627 - e30627

Published: May 1, 2024

Hepatotoxin carbon tetrachloride (CCl4) causes liver injury. This research aims to create ZnO-NPs using green synthesis from Moringa oleifera (MO) leaves aqueous extract, and chemically prepared confirming the by specialized equipment analysis. The sizes formed of were 80 55 nm for chemical methods, respectively. In addition, study their ability protect Wistar Albino male rats against oxidative stress exposed tetrachloride. MO leaf synthesized ZnO-NPs, at 100 200 mg/kg BW per day investigated hepatoprotective effects on enzyme biomarkers, renal antioxidant enzymes, lipid peroxidation, hematological parameters, histopathological changes. Compared control group, all kidney indicators considerably elevated after CCl4 injection. However, activity enzymes in was significantly reduced These outcomes indicate that greenly can restore normal function activity, as well enzymes. highest impact enhancing effect recorded received ZnO-NPs. increased drug delivery mechanism resulted a higher protective than extract.

Language: Английский

Citations

9
An evolutionary hybrid method based on particle swarm optimization algorithm and extreme gradient boosting for short-term streamflow forecasting DOI
Hüseyin Çağan Kılınç, Bülent Haznedar, Furkan Ozkan

et al.

Acta Geophysica, Journal Year: 2024, Volume and Issue: 72(5), P. 3661 - 3681

Published: Feb. 25, 2024

Language: Английский

Citations

8
An interpretable machine learning model for predicting 28-day mortality in patients with sepsis-associated liver injury DOI Creative Commons
Chengli Wen, Xu Zhang, Yong Li

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0303469 - e0303469

Published: May 20, 2024

Sepsis-Associated Liver Injury (SALI) is an independent risk factor for death from sepsis. The aim of this study was to develop interpretable machine learning model early prediction 28-day mortality in patients with SALI. Data the Medical Information Mart Intensive Care (MIMIC-IV, v2.2, MIMIC-III, v1.4) were used study. cohort MIMIC-IV randomized training set (0.7) and internal validation (0.3), MIMIC-III (2001 2008) as external validation. features more than 20% missing values deleted remaining multiple interpolated. Lasso-CV that lasso linear iterative fitting along a regularization path which best selected by cross-validation select important development. Eight models including Random Forest (RF), Logistic Regression, Decision Tree, Extreme Gradient Boost (XGBoost), K Nearest Neighbor, Support Vector Machine, Generalized Linear Models (CV_glmnet), Discriminant Analysis (LDA) developed. Shapley additive interpretation (SHAP) improve interpretability optimal model. At last, total 1043 included, whom 710 333 MIMIC-III. Twenty-four clinically relevant parameters construction. For SALI set, area under curve (AUC (95% CI)) RF 0.79 CI: 0.73–0.86), performed best. Compared traditional disease severity scores Oxford Acute Severity Illness Score (OASIS), Sequential Organ Failure Assessment (SOFA), Simplified Physiology II (SAPS II), Dysfunction (LODS), Systemic Inflammatory Response Syndrome (SIRS), III (APS III), also had performance. SHAP analysis found Urine output, Charlson Comorbidity Index (CCI), minimal Glasgow Coma Scale (GCS_min), blood urea nitrogen (BUN) admission_age five most affecting Therefore, has good predictive ability CCI, GCS_min, BUN age at admission(admission_age) within 24 h after intensive care unit(ICU) admission contribute significantly prediction.

Language: Английский

Citations

6
Optimized Deep Learning with Learning without Forgetting (LwF) for Weather Classification for Sustainable Transportation and Traffic Safety DOI Open Access
Surjeet Dalal, Bijeta Seth, Magdalena Rădulescu

et al.

Sustainability, Journal Year: 2023, Volume and Issue: 15(7), P. 6070 - 6070

Published: March 31, 2023

Unfortunately, accidents caused by bad weather have regularly made headlines throughout history. Some of the more catastrophic events to recently make news include a plane crash, ship collision, railway derailment, and several vehicle accidents. The public’s attention has been directed severe issue safety security under extreme conditions, many studies conducted highlight susceptibility transportation services environmental factors. An automated method determining weather’s state gained importance with development new technologies rise industry: intelligent transportation. Humans are well-suited for temperature from single photograph. Nevertheless, this is challenging problem fully autonomous system. objective research developing good classifier that uses only image as input. To resolve quality-of-life challenges, we propose modified deep-learning classify condition. proposed model based on Yolov5 model, which hyperparameter tuned Learning-without-Forgetting (LwF) approach. We took 1499 images Roboflow data repository divided them into training, validation, testing sets (70%, 20%, 10%, respectively). 99.19% accuracy. results demonstrated much higher accuracy level in comparison existing approaches. In future, may be implemented real-time.

Language: Английский

Citations

13
Random Forest Classifier Assessment on Liver Disease Estimating Through Smote-ENN Balancing for Precision and Complexity Matrix DOI

Pinamala Sruthi,

R. Suhasini,

V. Narasimha

et al.

Cognitive science and technology, Journal Year: 2025, Volume and Issue: unknown, P. 285 - 301

Published: Jan. 1, 2025

Language: Английский

Citations

0
Sodium Butyrate Alleviates the Toxic Effect of Aflatoxin B1 on Largemouth Bass (Micropterus Salmoides) DOI

Dongqiang Hou,

Hongxia Zhao, Kai Peng

et al.

Published: Jan. 1, 2025

Language: Английский

Citations

0
Uniform resource locator phishing detection using novel extreme gradient boosting algorithm in comparison with term frequency-inverse document frequency +N gram to improve accuracy DOI

Srinivas Islavath,

C. Rohith Bhat

AIP conference proceedings, Journal Year: 2025, Volume and Issue: 3252, P. 020057 - 020057

Published: Jan. 1, 2025

Language: Английский

Citations

0
Artificial intelligence for the noninvasive diagnosis of clinically significant portal hypertension DOI
Z. Z. Du, Ling Yang,

Hongliang He

et al.

Deleted Journal, Journal Year: 2025, Volume and Issue: 2(2), P. 100069 - 100069

Published: March 13, 2025

Language: Английский

Citations

0