Table 3.
Classifier, feature selection strategy and performance of the best lesion-wise models.
| Classifier | Selection | Train | Test | ||
|---|---|---|---|---|---|
| Accuracy | AUC | Accuracy | AUC | ||
| Radiomics | |||||
| GradientBoostingClassifier | RFE | 1.000 | 1.000 | 0.517 | 0.596 |
| ExtraTreesClassifier | RFA | 0.738 | 0.881 | 0.586 | 0.674 |
| AdaBoostClassifier | LASSO | 1.000 | 1.000 | 0.586 | 0.551 |
| Clinical scores | |||||
| SVC | RFE & RFA | 0.615 | 0.606 | 0.621 | 0.689 |
| SGDClassifier | LASSO | 0.615 | 0.618 | 0.621 | 0.649 |
| Combined features | |||||
| RandomForestClassifier | Best combined | 0.862 | 0.957 | 0.724 | 0.705 |
AUC, area under the curve; LASSO, least absolute shrinkage and selection operator; RFA, recursive feature addition; RFE, recursive feature elimination. See Supplementary Data S8 for more information.