Performance of machine learning models in predicting treatment response
Model | Risperidone | Aripiprazole | ||||
BS | AUROC | AUPRC | BS | AUROC | AUPRC | |
Backward stepwise LR | 0.124 | 0.659 | 0.244 | 0.117 | 0.672 | 0.300 |
LASSO | 0.122 | 0.678 | 0.253 | 0.115 | 0.681 | 0.304 |
Ridge | 0.123 | 0.680 | 0.255 | 0.117 | 0.683 | 0.307 |
e-net | 0.122 | 0.676 | 0.252 | 0.115 | 0.683 | 0.305 |
Random forest | 0.124 | 0.658 | 0.250 | 0.115 | 0.678 | 0.317 |
GBM | 0.121 | 0.686 | 0.269 | 0.114 | 0.682 | 0.314 |
SuperLearner | 0.121 | 0.685 | 0.264 | 0.115 | 0.688 | 0.314 |
AUPRC, area under the precision-recall curve; AUROC, area under the receiver operating characteristic curve; BS, Brier score; e-net, elastic net; GBM, gradient boosting machine; LASSO, least absolute shrinkage and selection operator; LR, logistic regression.