Skip to main content

Table 4 Comparison of the performance of the top 6 models in our study using Biomarker Scoring

From: Interpretable machine learning predicts cardiac resynchronization therapy responses from personalized biochemical and biomechanical features

Model Name

Accuracy

Train

Accuracy

Test

Recall

Train

Recall

Test

ROC-AUC

Test

F1

Test

MCC

Test

Voting Classifier

1.000

0.730

0.997

0.713

0.784

0.726

0.460

Stacking Classifier

0.855

0.723

0.915

0.800

0.772

0.744

0.451

Gradient Boosting Classifier

1.000

0.730

1.000

0.625

0.775

0.699

0.470

Logistic Regression

0.706

0.692

0.733

0.763

0.766

0.713

0.387

Random Forest Classifier

0.935

0.679

0.940

0.750

0.757

0.702

0.361

Adaptive Boosting Classifier

0.751

0.667

0.802

0.775

0.723

0.701

0.340