Table 2 Performance of gradually reduced expert set of variables for CBR data.
From: Decision making on vestibular schwannoma treatment: predictions based on machine-learning analysis
Variables | Train and validation (AVG, %) | Test (%) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Set | Number | ACC | PPV | TPR | TNR | AUC | ASE | ACC | PPV | TPR | TNR | AUC | ASE |
CBR | 38 | 78 | 75 | 84 | 72 | 87 | 15 | 72 | 69 | 79 | 65 | 81 | 19 |
CBREXPINI | 9 | 82 | 79 | 88 | 76 | 84 | 15 | 76 | 72 | 83 | 69 | 76 | 19 |
CBREXPFIN | 5 | 84 | 81 | 90 | 78 | 88 | 14 | 81 | 78 | 88 | 74 | 78 | 17 |
CBRCLASS | 38 | 86 | 84 | 91 | 82 | 92 | 11 | 88 | 85 | 93 | 82 | 87 | 12 |
