Development, validation, and visualization of a web-based nomogram for predicting chronic kidney disease incidence at health examination centers

Figures

• 

  Figure 1.. ROC curves. (a) ROC curves in the training cohort; (b) ROC curves in the internal validation cohort; (c) ROC curves in the external validation cohort. AUC: area under the curve; ROC: receiver operating characteristic.

• 

  Figure 2.. Decision curve analyses. (a) Decision curve analyses in the training cohort; (b) decision curve analyses in the internal validation cohort; (c) decision curve analyses in the external validation cohort.

• 

  Figure 3.. Nomogram for prediction of risk of developing CKD within 4 years in the healthy population. To use the nomogram, draw an upward vertical line from each variable axis to the ‘Points’ bar to calculate the corresponding points for each variable. Then sum the covariate points and draw a downward vertical line from the ‘Total Points’ bar to obtain the predictive probability of developing CKD within 4 years. The higher the total points, the higher the risk of CKD. Alb: albumin; CKD: chronic kidney disease; DM: diabetes mellitus; SBP: systolic blood pressure; Scr: serum creatinine; TG: serum triglyceride.

• 

  Figure 4.. The intuitive interface of the web-based nomogram (https://luochuxuan.shinyapps.io/dynnomapp/). Instruction for using the web-based nomogram: Enter model parameter values in the left input box and click the ‘Predict’ button below to start the analysis. The corresponding individual probability of developing CKD within 4 years can be queried in the ‘Graphical Summary’ and ‘Numerical Summary’ modules on the right side of the interface. CKD: chronic kidney disease; Alb: albumin; DM: diabetes mellitus; SBP: systolic blood pressure; Scr: serum creatinine; TG: serum triglyceride.