Related Experiment Video
Updated: Apr 30, 2026

Robot-Assisted Kidney Transplantation
Published on: July 19, 2021
A nomogram for predicting renal function recovery after robotic-assisted ureteral reconstruction: development and
Mu-Yang Xu1,2,3, Bin-Bin Gong1,2,3, Yu He1,2,3
1¹Department of Urology, The First Affiliated Hospital of Anhui Medical University, Anhui Medical University, No. 218 Jixi Road, Hefei, Anhui, P.R. China.
A simple 4-variable model accurately predicts kidney function recovery after ureteric obstruction surgery. This clinical nomogram balances performance and practicality for patient counseling and surgical decisions.
Area of Science:
- Urology
- Nephrology
- Medical Informatics
- Machine Learning in Medicine
Background:
- Ureteric obstruction can impair renal function.
- Predicting renal recovery post-surgical repair is crucial for patient management.
- Developing accurate predictive models aids clinical decision-making.
Purpose of the Study:
- To develop and validate predictive models for significant renal function recovery after ureteric obstruction repair.
- To compare a traditional Multivariable Logistic Regression model with a Machine Learning-based LASSO Regression Model.
- To present the most practical model as a clinical nomogram for predicting renal recovery.
Main Methods:
- Retrospective analysis of 100 adult patients undergoing surgical repair for unilateral ureteric obstruction.
- Development of a Multivariable Model (backward stepwise regression) and a LASSO Model (least absolute shrinkage and selection operator regression).
- Performance evaluation using discrimination (AUC), calibration (calibration curve), and clinical utility (Decision Curve Analysis), with internal validation via bootstrap resampling.
Main Results:
- The Multivariable Model identified four predictors: age, preoperative ipsilateral GFR, renal atrophy, and SFU Grade 4.
- The LASSO model selected 12 variables, showing slightly higher discrimination (AUC 0.841 vs. 0.827) and validated AUC (0.779 vs. 0.776).
- Despite marginal statistical advantages of the LASSO model, the 4-variable Multivariable Model offered superior balance of accuracy, interpretability, and clinical utility, leading to its selection for nomogram development.
Conclusions:
- A robust 4-variable nomogram was developed for predicting renal recovery after ureteric obstruction repair.
- The simpler Multivariable Model demonstrated excellent calibration and comparable clinical utility to the more complex LASSO model.
- The developed nomogram serves as a valuable tool for patient counseling and surgical decision-making in cases of ureteric obstruction.
More Related Videos
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
05:34A Mouse Model to Evaluate the Long-Term Structural and Functional Outcomes after the Reversal of Prolonged Unilateral Ureteric Obstruction
Published on: July 18, 2025