Abstract
ObjectiveThis study sought to develop a prediction model for double-J pipe scab following radical total cystectomy.MethodsThis retrospective study included clinical data on 175 patients who underwent routine double-J pipe replacement after radical total cystectomy. A double-J pipe scab was the nomogram's outcome. Boruta feature selection and the least absolute shrinkage selection operator (LASSO) approach were employed to predictions as efficiently as possible. Using multiple logistic regression, predictive models were created and displayed as nomograms. Nomogram performance was assessed using decision curve analysis, calibration plots, and receiver operating characteristic (ROC) curves. By computing the validation cohort's performance, the model was internally validated.ResultsThis study covered 175 patients in total. In twenty-nine individuals (16.57%), a double-J pipe scab formed. Every participant was randomly split into two groups: training (n = 122) and validation (n = 53). As predictors, this nomogram included urine leukocytes, urinary PH, daily water intake, BMI, and double-J pipe brand. Excellent identification performance is indicated by the training and verification groups' ROC curves, while the calibration curves demonstrate both groups' excellent correction outcomes.ConclusionThis study establishes a foundation for preventing and treating double-J pipe stones. It has a substantial predictive value for the occurrence of double-J pipe stones in the double-J pipe following radical total cystectomy.