TY - JOUR
T1 - Predicting intra-operative and postoperative consequential events using machine-learning techniques in patients undergoing robot-assisted partial nephrectomy
T2 - a Vattikuti Collective Quality Initiative database study
AU - Bhandari, Mahendra
AU - Nallabasannagari, Anubhav Reddy
AU - Reddiboina, Madhu
AU - Porter, James R.
AU - Jeong, Wooju
AU - Mottrie, Alexandre
AU - Dasgupta, Prokar
AU - Challacombe, Ben
AU - Abaza, Ronney
AU - Rha, Koon Ho
AU - Parekh, Dipen J.
AU - Ahlawat, Rajesh
AU - Capitanio, Umberto
AU - Yuvaraja, Thyavihally B.
AU - Rawal, Sudhir
AU - Moon, Daniel A.
AU - Buffi, Nicolò M.
AU - Sivaraman, Ananthakrishnan
AU - Maes, Kris K.
AU - Porpiglia, Francesco
AU - Gautam, Gagan
AU - Turkeri, Levent
AU - Meyyazhgan, Kohul Raj
AU - Patil, Preethi
AU - Menon, Mani
AU - Rogers, Craig
N1 - Funding Information:
We gratefully acknowledge discussions and comments on the manuscript by our colleague Trevor Zeffiro. We are grateful to the Vattikuti Foundation for granting access to the VCQI database and RediMinds for funding this work. This publication only reflects the authors views. The funding agency is not liable for any use that may be made of the information contained herein.
Funding Information:
We gratefully acknowledge discussions and comments on the manuscript by our colleague Trevor Zeffiro. We are grateful to the Vattikuti Foundation for granting access to the VCQI database and RediMinds for funding this work. This publication only reflects the authors views. The funding agency is not liable for any use that may be made of the information contained herein.
Publisher Copyright:
© 2020 The Authors BJU International © 2020 BJU International Published by John Wiley & Sons Ltd
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/9/1
Y1 - 2020/9/1
N2 - Objective: To predict intra-operative (IOEs) and postoperative events (POEs) consequential to the derailment of the ideal clinical course of patient recovery. Materials and Methods: The Vattikuti Collective Quality Initiative is a multi-institutional dataset of patients who underwent robot-assisted partial nephectomy for kidney tumours. Machine-learning (ML) models were constructed to predict IOEs and POEs using logistic regression, random forest and neural networks. The models to predict IOEs used patient demographics and preoperative data. In addition to these, intra-operative data were used to predict POEs. Performance on the test dataset was assessed using area under the receiver-operating characteristic curve (AUC-ROC) and area under the precision-recall curve (PR-AUC). Results: The rates of IOEs and POEs were 5.62% and 20.98%, respectively. Models for predicting IOEs were constructed using data from 1690 patients and 38 variables; the best model had an AUC-ROC of 0.858 (95% confidence interval [CI] 0.762, 0.936) and a PR-AUC of 0.590 (95% CI 0.400, 0.759). Models for predicting POEs were trained using data from 1406 patients and 59 variables; the best model had an AUC-ROC of 0.875 (95% CI 0.834, 0.913) and a PR-AUC 0.706 (95% CI, 0.610, 0.790). Conclusions: The performance of the ML models in the present study was 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.
AB - Objective: To predict intra-operative (IOEs) and postoperative events (POEs) consequential to the derailment of the ideal clinical course of patient recovery. Materials and Methods: The Vattikuti Collective Quality Initiative is a multi-institutional dataset of patients who underwent robot-assisted partial nephectomy for kidney tumours. Machine-learning (ML) models were constructed to predict IOEs and POEs using logistic regression, random forest and neural networks. The models to predict IOEs used patient demographics and preoperative data. In addition to these, intra-operative data were used to predict POEs. Performance on the test dataset was assessed using area under the receiver-operating characteristic curve (AUC-ROC) and area under the precision-recall curve (PR-AUC). Results: The rates of IOEs and POEs were 5.62% and 20.98%, respectively. Models for predicting IOEs were constructed using data from 1690 patients and 38 variables; the best model had an AUC-ROC of 0.858 (95% confidence interval [CI] 0.762, 0.936) and a PR-AUC of 0.590 (95% CI 0.400, 0.759). Models for predicting POEs were trained using data from 1406 patients and 59 variables; the best model had an AUC-ROC of 0.875 (95% CI 0.834, 0.913) and a PR-AUC 0.706 (95% CI, 0.610, 0.790). Conclusions: The performance of the ML models in the present study was 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.
KW - deep learning
KW - intra-operative complications
KW - machine learning
KW - postoperative complications
KW - postoperative morbidity
KW - robot-assisted partial nephrectomy
UR - http://www.scopus.com/inward/record.url?scp=85084839276&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85084839276&partnerID=8YFLogxK
U2 - 10.1111/bju.15087
DO - 10.1111/bju.15087
M3 - Article
C2 - 32315504
AN - SCOPUS:85084839276
VL - 126
SP - 350
EP - 358
JO - BJU International
JF - BJU International
SN - 1464-4096
IS - 3
ER -