Table 5. Comparison of the N-SRS, and the R-FSRS performance on the Validation population using the c-statistic.
BMC–White | BMC–Black | BMC–Hispanic | ||||
---|---|---|---|---|---|---|
Model | AUC | 95% CI | AUC | 95% CI | AUC | 95% CI |
N-SRS | 74.30% | 0.7149–0.771 | 75.80% | 0.7345–0.767 | 72.79% | 0.6889–0.7671 |
R-FSRS (both genders) | 64.91% | 0.6266–0.6716 | 64.85% | 0.6304–0.6666 | 61.04% | 0.5601–0.6587 |
R-FSRS (women) | 67.35% | 0.6411–0.7058 | 65.22% | 0.628–0.6764 | 61.06% | 0.5548–0.6663 |
R-FSRS (men) | 62.46% | 0.5931–0.6555 | 64.49% | 0.6181–0.6717 | 61.01% | 0.5621–0.6587 |
Log.Reg | 71.55% | 0.6823–0.7402 | 69.77% | 0.6823–0.7402 | 70.46% | 0.6765–0.7359 |
CART | 69.01% | 0.6627–0.7134 | 66.41% | 0.6272–0.6609 | 66.10% | 0.6286–0.6934 |
Random Forest | 75.08% | 0.7162–0.7855 | 73.14% | 0.7139–0.749 | 70.80% | 0.6807–0.7354 |
XGBoost | 77.32% | 0.7582–0.7881 | 74.88% | 0.7133–0.7842 | 74.27% | 0.7187–0.7667 |