Development and validation of a diagnostic model for mild cognitive impairment in patients with type 2 diabetes mellitus and concomitant white matter hyperintensities

To construct and validate a nomogram model that effectively discriminates mild cognitive impairment (MCI) from normal cognition in patients with type 2 diabetes mellitus (T2DM) and concomitant white matter hyperintensities (WMH). This single‑center cross-sectional study included 1,175 consecutive T2DM patients with WMH from Cangzhou Central Hospital (January 2020 – December 2025), randomly assigned to a training cohort ( n = 823) and a validation cohort ( n = 352) in a 7:3 ratio. LASSO regression and multivariable logistic regression with four adjustment models were used to select associated variables. A nomogram was constructed and evaluated by receiver operating characteristic curves (ROC), calibration plots, and decision curve analysis (DCA). Eight independent associated variables were identified: C-peptide, HbA1c, mean amplitude of glycemic excursions (MAGE), coefficient of variation (CV), time in range (TIR), Fazekas scale, weight-adjusted waist index (WWI), and urinary albumin-to-creatinine ratio (UACR). The nomogram showed good discrimination in the training cohort [area under the curve (AUC) = 0.813, 95%CI 0.782–0.843] and validation cohort (AUC = 0.773, 95%CI 0.720–0.827). Calibration was excellent (training: intercept 0.00, slope 1.00; Hosmer-Lemeshow P = 0.317; validation: intercept 0.13, slope 0.90; P = 0.432). DCA confirmed clinical net benefit. This novel nomogram, incorporating glycemic variability, central obesity, insulin resistance, renal-vascular injury, and WMH burden, accurately discriminates the presence of MCI in T2DM patients with WMH, offering a practical tool for early risk stratification and intervention.