Table 5 Test AUROCs for XGBoost for predicting 1 year mortality using all features or pre-transplant features only stratified by transplant type.

From: In silico perturbations provide multivariate interpretability in predicting post-lung transplant outcomes

 

All features

Pre-tx features

Both

0.63 [0.60, 0.67] 95% CI

0.60 [0.57,0.64] 95% CI

Single transplants

0.67 [0.61, 0.71] 95% CI

0.58 [0.52,0.63] 95% CI

Bilateral transplants

0.57 [0.52, 0.62] 95% CI

0.55 [0.50,0.60] 95% CI

Back to article page