AI-driven Prediction of Insulin Resistance in Normal Populations: Comparing Models and Criteria

2025-03-07

AI Summary

Early identification of insulin resistance (IR) in non-diabetic populations remains challenging due to reliance on invasive, costly, or multi-parameter assays. Method: We developed a lightweight, interpretable AI model using only routinely available clinical measurements—primarily fasting plasma glucose (FPG) and waist circumference—trained via CatBoost and rigorously validated across independent cohorts (NHANES and CHARLS). Contribution/Results: To our knowledge, this is the first high-accuracy IR prediction model driven solely by a single glycemic marker (FPG). SHAP analysis identified waist circumference as the most influential interpretable feature, eliminating dependence on multiple blood biomarkers. Under the METS-IR criterion, the model achieved AUCs of 0.9731 (internal) and 0.9591 (external), with RMSE = 3.057; under the TyG index, AUCs were 0.7777 and 0.7442—significantly outperforming existing simplified models. This non-invasive, cost-effective, and clinically interpretable tool supports primary prevention of type 2 diabetes and cardiovascular disease.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Hardware-aware ML
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Insulin resistance (IR) is a key precursor to diabetes and a significant risk factor for cardiovascular disease. Traditional IR assessment methods require multiple blood tests. We developed a simple AI model using only fasting blood glucose to predict IR in non-diabetic populations. Data from the NHANES (1999-2020) and CHARLS (2015) studies were used for model training and validation. Input features included age, gender, height, weight, blood pressure, waist circumference, and fasting blood glucose. The CatBoost algorithm achieved AUC values of 0.8596 (HOMA-IR) and 0.7777 (TyG index) in NHANES, with an external AUC of 0.7442 for TyG. For METS-IR prediction, the model achieved AUC values of 0.9731 (internal) and 0.9591 (external), with RMSE values of 3.2643 (internal) and 3.057 (external). SHAP analysis highlighted waist circumference as a key predictor of IR. This AI model offers a minimally invasive and effective tool for IR prediction, supporting early diabetes and cardiovascular disease prevention.

Problem

• AI model predicts insulin resistance using fasting blood glucose.
• Model validated with NHANES and CHARLS datasets for accuracy.
• Minimally invasive tool aids early diabetes and cardiovascular prevention.

Innovation

• AI model predicts insulin resistance non-invasively
• Uses fasting glucose, demographics, and body metrics
• Achieves high accuracy with CatBoost algorithm