Retrospective Evaluation of Bayesian Risk Models of LVAD Mortality at a Single Implant Center.

摘要: Use of a left ventricular assist device (LVAD) can benefit patients with end stage heart failure, but only careful patient selection. In this study, previously derived Bayesian network models for predicting LVAD mortality at 1, 3, and 12 months post-implant were evaluated on retrospective data from single implant center. The performed well all three time points, receiver operating characteristic area under the curve (ROC AUC) 78, 76, 75%, respectively. This evaluation model performance verifies utility these in “real life” scenarios an individual institution.