Ensemble Prediction of Time to Event Outcomes with Competing Risks: A Case Study of Surgical Complications in Crohn's Disease
This work addresses surgical complication prediction for Crohn's disease patients, representing an incremental domain-specific advancement.
The authors tackled the problem of predicting major abdominal surgery within 5 years after Crohn's disease diagnosis using baseline covariates, achieving a method that extends machine learning to right-censored data with proposed performance metrics.
We develop a novel algorithm to predict the occurrence of major abdominal surgery within 5 years following Crohn's disease diagnosis using a panel of 29 baseline covariates from the Swedish population registers. We model pseudo-observations based on the Aalen-Johansen estimator of the cause-specific cumulative incidence with an ensemble of modern machine learning approaches. Pseudo-observation pre-processing easily extends all existing or new machine learning procedures to right-censored event history data. We propose pseudo-observation based estimators for the area under the time varying ROC curve, for optimizing the ensemble, and the predictiveness curve, for evaluating and summarizing predictive performance.