Fig. 4

ROC curve and relative importance of VAII, VAI and CRP for predicting stroke. (A) Receiver Operating Characteristic (ROC) curves for VAII, VAI, and CRP were used independently to evaluate their discriminatory ability by calculating the Area Under the Curve (AUC). (B) The base model included the following variables: Age, eGFR (Cystatin C), Depression, Gender, Marital Status, Education Level, Rural Area, Smoking Status, Diabetes, Hypertension, Antidiabetic Medication, Antihypertensive Medication, Drinking Status, Dyslipidemia, Loneliness, Exercise Status, and Heart Disease. Combine VAII, VAI and CRP with the basic model in that order. (C) Relative importance of risk factors for predicting stroke.