Personalized application of machine learning algorithms to identify pediatric patients at risk for recurrent ureteropelvic junction obstruction after dismembered pyeloplasty
Erik Drysdale, MSc, Adree Khondker, BSc, Jin Kyu Kim, MD, Jethro CC Kwong, MD, Lauren Erdman, PhD(c), Michael Chua, MD, Daniel T. Keefe, MD, Marisol Lolas, MD, Joana Dos Santos, MD, Gregory E. Tasian, MD, Mandy Rickard, MN, NP-Pediatrics, Armando J. Lorenzo, MD.
SickKids, Toronto, ON, Canada.
BACKGROUND: Pyeloplasty is the mainstay surgical intervention to manage ureteropelvic junction obstruction (UPJO) in children. Although pyeloplasty has a high success rate, between 5 and 10% of children will require re-intervention due to recurrent obstruction. Although several factors can explain the risk of recurrence, it is challenging to predict which patients, individually, are at the highest risk for this in the postoperative course. The objectives of this study were to develop a model that predicts whether a child will develop a recurrent obstruction after pyeloplasty, determine their survival risk score, and expected time to re-intervention using machine learning (ML).
METHODS: We reviewed patients undergoing pyeloplasty from 2008 to 2020 at our institution, including all children and adolescents younger than 18 years. We developed a two-stage machine learning model from 34 clinical fields (Figure 1), which included patient characteristics, ultrasound findings, and anatomical variation. We fit and trained with a logistic lasso model for binary cure model and subsequent survival model. Feature importance on the model was determined with post-selection inference. Performance metrics included area under the receiver-operating-characteristic (AUROC), concordance, and leave-one-out cross validation.
RESULTS: A total of 543 patients were identified, with a median preoperative and postoperative anteroposterior diameter of 23 mm and 10 mm, respectively. 39 of 232 patients included in the survival model required re-intervention. The cure and survival models performed well with a leave-one-out cross validation AUROC and concordance of 0.86 and 0.78, respectively. Post-selective inference showed that larger anteroposterior diameter at the second post-op follow up, and anatomical variation in the form of concurrent anomalies were significant model features predicting negative outcomes. For example, the predicted likelihood of re-intervention for two patients with low (6 mm) vs high (30 mm) post-operative anteroposterior diameter is provided (Figure 2). The model can be used at https://sickkidsurology.shinyapps.io/PyeloplastyReOpRisk/.
CONCLUSIONS:
Conclusion: Our ML-based model performed well in predicting the risk of and time to re-intervention after pyeloplasty. The implementation of this ML-based approach is novel in pediatric urology, and will likely help achieve point-of-care personalized risk-stratification for patients. Further real-world validation is warranted. 
Back to 2021 Abstracts