This study presents a machine learning model to predict renal function decline following minimally-invasive partial nephrectomy. Using a dataset of 556 patients treated between 2015 and 2023, the model incorporated patient, tumor, and intraoperative surgical variables – including clamping strategy, resection technique, and renorrhaphy type – to estimate the 3-month postoperative eGFR drop. A Random Forest Regressor outperformed other models, achieving a prediction accuracy of 89.29%, a mean absolute error of 8.09 mL/min/1.73 m2, and a strong correlation with observed outcomes (r=0.904, P<10−42). These findings support the use of AI for personalized surgical planning and functional outcome prediction in nephron-sparing surgery.
From planning to prognosis: predicting renal function after minimally-invasive partial nephrectomy with artificial intelligence
Amparore D.;Pezzi V.;Di Dio M.;Fiori C.;Porpiglia F.
2025-01-01
Abstract
This study presents a machine learning model to predict renal function decline following minimally-invasive partial nephrectomy. Using a dataset of 556 patients treated between 2015 and 2023, the model incorporated patient, tumor, and intraoperative surgical variables – including clamping strategy, resection technique, and renorrhaphy type – to estimate the 3-month postoperative eGFR drop. A Random Forest Regressor outperformed other models, achieving a prediction accuracy of 89.29%, a mean absolute error of 8.09 mL/min/1.73 m2, and a strong correlation with observed outcomes (r=0.904, P<10−42). These findings support the use of AI for personalized surgical planning and functional outcome prediction in nephron-sparing surgery.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
