Machine learning-based prediction of stone-free rate after retrograde intrarenal surgery for lower pole renal stones

1

Lower pole stones (LPS) account for 25-35% of renal stones and present unique challenges for retrograde intrarenal surgery (RIRS).
2

The stone-free rate (SFR) for LPS remains lower than for stones in other locations due to anatomical challenges like acute infundibulopelvic angle.
3

Machine learning (ML) models were developed to predict SFR after RIRS by analyzing various patient and stone-related characteristics.
4

The study included 327 patients, with data split for model development and validation, ensuring generalizability of the results.
5

SHAP values were used to interpret feature importance in the ML model, highlighting factors positively or negatively correlated with stone-free status.

World Journal Of Urology

by Hsiang Ying Lee, Yu-Hung Tung, Jose Carlo Elises, Yen-Chun Wang, Vineet Gauhar, Sung Yong Cho
July 12, 2025

1