Predicting insulin resistance in non-diabetics: Comparing obesity indicators using machine learning.
- 1Division of Endocrinology and Metabolism, Department of Internal Medicine, Nantong Third People's Hospital, Nantong University, Nantong, Jiangsu, China.
Abstract
This cross-sectional study aimed to compare obesity indicators for predicting insulin resistance using machine learning. Data were obtained from NHANES 2015 to 2018, focusing on a nondiabetic adult population. Insulin resistance was defined as HOMA-IR ≥ 75th percentile of the training set. Key predictors were selected via bootstrap stability selection with elastic net. Logistic regression, random forest, and extreme gradient boosting (XGBoost) models were built using body mass index (BMI) and waist-to-height ratio (WHtR). Performance was assessed by area under the curve (AUC), sensitivity, specificity, accuracy, calibration, and decision curve analysis. Race-stratified and nested regression analyses were performed. A total of 3562 subjects were included. BMI, WHtR, Glycated hemoglobin, and triglycerides were identified as key predictors. All models achieved AUC > 0.79; logistic regression (LR) performed best (BMI-based LR: AUC = 0.811; WHtR-based LR: AUC = 0.811). WHtR-based LR showed slightly better accuracy (0.746 vs 0.715) and specificity (0.744 vs 0.678). Obesity indicators had the highest feature importance, and race-stratified analyses confirmed general robustness across subgroups. WHtR and BMI are strong independent predictors of IR. LR models using these simple indices performed well in discrimination, calibration, and net benefit, offering cost-effective screening options, particularly where resources are scarce. Future work should prioritize external validation and race-specific thresholds.
Related
- Predicting the Risk of Distal Junctional Kyphosis After Surgery for Adolescent Idiopathic Scoliosis Using Preoperative Radiographic Indicators and Machine Learning
- Multi-model comparison and nomogram development for predicting in-hospital mortality in Trauma patients based on NLR and other indicators
- Comparing physical activity policy evaluations across 27 EU member states: Diverse indicators, inconsistent findings
- Comparing nutrition indicators in adult population monitoring systems in Italy, Estonia and Belgium
- Exercise and the Kynurenine pathway: Current state of knowledge and results from a randomized cross-over study comparing acute effects of endurance and resistance training
BibTeX
@article{guo2026predicting,
title = {{Predicting insulin resistance in non-diabetics: Comparing obesity indicators using machine learning.}},
author = {Zhiyu Guo and Yingfeng Wei and Xuechao Lu and Chunhua Wang},
journal = {Medicine},
year = {2026},
month = oct,
doi = {10.1097/md.0000000000051034},
url = {https://europepmc.org/article/MED/42888588},
note = {CC BY},
}
RIS
TY - JOUR
TI - Predicting insulin resistance in non-diabetics: Comparing obesity indicators using machine learning.
AU - Zhiyu Guo
AU - Yingfeng Wei
AU - Xuechao Lu
AU - Chunhua Wang
PY - 2026
DA - 2026/10/01
JO - Medicine
DO - 10.1097/md.0000000000051034
UR - https://europepmc.org/article/MED/42888588
AB - This cross-sectional study aimed to compare obesity indicators for predicting insulin resistance using machine learning. Data were obtained from NHANES 2015 to 2018, focusing on a nondiabetic adult population. Insulin resistance was defined as HOMA-IR ≥ 75th percentile of the training set. Key predictors were selected via bootstrap stability selection with elastic net. Logistic regression, random forest, and extreme gradient boosting (XGBoost) models were built using body mass index (BMI) and waist-to-height ratio (WHtR). Performance was assessed by area under the curve (AUC), sensitivity, specificity, accuracy, calibration, and decision curve analysis. Race-stratified and nested regression analyses were performed. A total of 3562 subjects were included. BMI, WHtR, Glycated hemoglobin, and triglycerides were identified as key predictors. All models achieved AUC > 0.79; logistic regression (LR) performed best (BMI-based LR: AUC = 0.811; WHtR-based LR: AUC = 0.811). WHtR-based LR showed slightly better accuracy (0.746 vs 0.715) and specificity (0.744 vs 0.678). Obesity indicators had the highest feature importance, and race-stratified analyses confirmed general robustness across subgroups. WHtR and BMI are strong independent predictors of IR. LR models using these simple indices performed well in discrimination, calibration, and net benefit, offering cost-effective screening options, particularly where resources are scarce. Future work should prioritize external validation and race-specific thresholds.
SN - 0304-5412
SN - 1536-5964
ER -