Table 6 Detailed metrics for the three best performing classifiers for CBR data.
From: Decision making on vestibular schwannoma treatment: predictions based on machine-learning analysis
Classification (feature selection) | Train and validation (avg, %) | Test (%) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
ACC | PPV | TPR | TNR | AUC | ASE | ACC | PPV | TPR | TNR | AUC | ASE | |
Neural network (Gradient boosting) | 84 | 82 | 88 | 80 | 93 | 10 | 89 | 87 | 93 | 85 | 92 | 8 |
Gradient boosting (CBREXPFIN) | 84 | 81 | 89 | 79 | 89 | 13 | 87 | 84 | 94 | 80 | 86 | 15 |
Gradient boosting (random forest) | 91 | 89 | 96 | 86 | 94 | 10 | 87 | 84 | 93 | 81 | 84 | 14 |
