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

TL;DR

A simple AI model using only fasting blood glucose to predict IR in non-diabetic populations offers a minimally invasive and effective tool for IR prediction, supporting early diabetes and cardiovascular disease prevention.

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.

Figures (8)

• Figure 1: Flow diagram of study population selection process.
• Figure 2: ROC curves of HOMA-IR classification using different methods. Due to the lack of Plasma insulin levels in the CHARLS dataset, external validation calculations cannot be performed.
• Figure 3: ROC curves of TyG classification using different methods. a) NHANES test set results. b) CHARLS external test results.
• Figure 4: ROC curves of METS-IR classification using different methods. a) NHANES test set results. b) CHARLS external test results.
• Figure 5: Scatter plot images of METS-IR numerical prediction using TabKANet. a) NHANES test set results. b) CHARLS external test results.