Table 5 Responsiveness and reliability in predicting mortality for the 10 models developed using all 15 variables
TKA Mortality | ||||||
|---|---|---|---|---|---|---|
| Â | Reliability (Accuracy) | Responsiveness (AUC) | ||||
| Â | Training | Testing | Validation | Training | Testing | Validation |
Random Forest | 93.49% | 93.49% | 93.47% | 0.941 | 0.687 | 0.749 |
Neural Network | 93.47% | 93.49% | 93.47% | 0.816 | 0.938 | 0.996 |
XGT Boost Tree | 99.97% | 99.97% | 99.98% | 0.921 | 0.839 | 0.954 |
XGT Boost linear | 99.97% | 99.97% | 99.81% | 0.982 | 0.938 | 0.997 |
LSVM | 99.87% | 99.89% | 99.89% | 0.981 | 0.944 | 0.997 |
CHAID | 99.87% | 99.89% | 99.89% | 0.978 | 0.901 | 0.991 |
Decision List | 99.90% | 99.90% | 99.91% | 0.845 | 0.925 | 0.871 |
Discriminant | 86.61% | 86.72% | 86.39% | 0.894 | 0.97 | 0.93 |
Logistic Regression | 93.21% | 93.26% | 93.17% | 0.86 | 0.932 | 0.996 |
Bayesian Network | 93.47% | 93.49% | 93.47% | 0.931 | 0.821 | 0.632 |