Predicting intraoperative and postoperative consequential events using machine learning techniques in patients undergoing robotic partial nephrectomy (RPN): Vattikuti Collective Quality Initiative (VCQI) database study.

Document Type


Publication Date


Publication Title

BJU international


OBJECTIVE: To predict intraoperative events (IOE) and postoperative events (POE) consequential to the derailment of the ideal clinical course of patient recovery.

MATERIAL AND METHODS: Vattikuti Collective Quality Initiative (VCQI), a multi-institutional dataset of patients who underwent Robotic Partial Nephrectomy for kidney tumors. Machine Learning (ML) models were constructed to predict IOE, and POE using Logistic Regression, Random Forest, and Neural Networks. The models to predict IOE used patient demographics and preoperative data. In addition to the above, intraoperative data was used to predict POE. Performance on the test dataset was assessed using Area Under Receiver Operating Curve (AUC-ROC) and Area Under Precision-Recall Curve (PR-AUC).

RESULTS: The rate of IOE and POE was 5.62% and 20.98%, respectively. Models for predicting IOE were constructed using data from 1690 patients and 38 variables; the best model had AUC-ROC of 0.858 (95% CI, 0.762, 0.936), and PR-AUC of 0.590 (95% CI, 0.400, 0.759). Models for predicting POE were trained using data from 1406 patients and 59 variables; the best model had AUC-ROC of 0.875 (95% CI, 0.834, 0.913), and PR-AUC 0.706 (95% CI, 0.610, 0.790).

CONCLUSIONS: The performance of the ML models in this study is encouraging. Further validation in a multi-institutional clinical setting with larger datasets would be necessary to establish their clinical value. ML models can be used to predict significant events during and after surgery with good accuracy, paving the way for application in clinical practice to predict and intervene at an opportune time to avert complications and improve patient outcomes.

Clinical Institute

Kidney & Diabetes