What question did this study set out to answer?

This study aims to enhance diabetes risk prediction using machine learning methods applied to population-level data.

August 11, 2026Scientific ReportsOpen Access

Machine learning-based classification of diabetes mellitus using sociodemographic, behavioral, and clinical predictor

View Full Paper

Ask AI

Bookmark

Authors

FAFetlework Gubena Arage TTTigist Kifle Tsegaw TATsegasilassie Gebremariam Abate

Discussion

Member takes

Overview

Randomized trial develops machine learning models for diabetes risk prediction in low-income populations, indicating improved early detection potential.

Key Points

This study aims to enhance diabetes risk prediction using machine learning methods applied to population-level data.
Analyzed data from WHO STEPS surveys covering 56,560 adults across 14 low- and lower-middle-income countries.
Used various machine learning algorithms and preprocessing techniques including SMOTEENN to handle class imbalance.
Evaluated model performance using accuracy, recall, AUC, and SHAP values for feature interpretation.
XGBoost outperformed other models with 90% accuracy, 87% balanced accuracy, and 94% AUC.
Confusion matrix for XGBoost indicated strong discrimination with few misclassifications among diabetic and non-diabetic individuals.
Top predictors identified included cholesterol, age, residency, tobacco use, and waist circumference.

Cite This Study

Arage et al. (2026) studied this question.

synapsesocial.com/papers/6a7ace0a3401087f2249dbdc https://doi.org/10.1038/s41598-026-64743-x

View Full Paper

Ask AI

Bookmark

August 11, 2026Scientific ReportsOpen Access

Machine learning-based classification of diabetes mellitus using sociodemographic, behavioral, and clinical predictor

View Full Paper

Ask AI

Bookmark

Authors

FAFetlework Gubena Arage TTTigist Kifle Tsegaw TATsegasilassie Gebremariam Abate

Discussion

Member takes

Overview

Randomized trial develops machine learning models for diabetes risk prediction in low-income populations, indicating improved early detection potential.

Key Points

This study aims to enhance diabetes risk prediction using machine learning methods applied to population-level data.
Analyzed data from WHO STEPS surveys covering 56,560 adults across 14 low- and lower-middle-income countries.
Used various machine learning algorithms and preprocessing techniques including SMOTEENN to handle class imbalance.
Evaluated model performance using accuracy, recall, AUC, and SHAP values for feature interpretation.
XGBoost outperformed other models with 90% accuracy, 87% balanced accuracy, and 94% AUC.
Confusion matrix for XGBoost indicated strong discrimination with few misclassifications among diabetic and non-diabetic individuals.
Top predictors identified included cholesterol, age, residency, tobacco use, and waist circumference.

Cite This Study

Arage et al. (2026) studied this question.

synapsesocial.com/papers/6a7ace0a3401087f2249dbdc https://doi.org/10.1038/s41598-026-64743-x

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark