Abstract

Diabetic foot ulcers are a severe complication of type 2 diabetes mellitus, but the pathophysiology behind diabetic foot formation in response to insulin levels is unclear. The objective of the present study was to investigate the correlation between insulin levels and diabetic foot ulcers among patients with type 2 diabetes, using sound statistical and machine learning analysis. Cross-sectional analysis was conducted among 136 type 2 diabetic patients, including 34 controls with no foot complications and 102 with diabetic foot ulcers. Clinical parameters of fasting insulin, indices of insulin resistance, demographic information, and metabolic markers were analyzed by statistical correlations, t-tests, and machine learning models (Random Forest, Logistic Regression) to ascertain predictive association. There was a significant positive correlation (r = 0.354, p < 0.05) between insulin concentration and diabetic foot ulcer presence. Mean insulin concentrations were significantly higher in diabetic foot ulcer patients (8.04 ± 4.2 μU/mL) compared to controls (5.77 ± 2.1 μU/mL). Machine learning model analysis found insulin concentration to be the most significant predictor, with an area under the curve of 0.762, indicating good discriminative potential for clinical screening. Elevated insulin levels show a significant association with diabetic foot ulcers in type 2 diabetes, suggesting that the measurement of insulin could be an effective biomarker for risk stratification and prevention at an early level.