Table 6 Comparison with similar works from the literature
From: Machine learning-based prognostic model for 30-day mortality prediction in Sepsis-3
Authors | Approach | Evaluation |
|---|---|---|
Hou et al. [41] | Logistic regression model, SAPS-II scores prediction model and XGBoost algorithm | Logistic regression (AUC: 0.819), SAPS-II (AUC: 0.797) and XGBoost (AUC: 0.857) |
Su et al. [14] | ANN-based architecture | AUC of 0.873 |
Park et al. [38] | Logistic Regression, Random Forest, XGBoost, Deep Neural Network (DNN), Super Learner Model | Logistic Regression (AUC: 0.878), Random Forest (AUC: 0.878), XGBoost (AUC: 0.888), DNN (AUC: 0.893), Super Learner Model (AUC: 0.883) |
Yang et al. [22] | Logistic Regression Analysis | Training & Validation AUROC of 0.763 and 0.753 |
Palmowski et al. [75] | SVM with polynomial kernel, ANN | SVM (AUC: 0.84), ANN (AUC: 0.82) |
Zhang et al. [65] | XGBoost | AUC: 0.94, F1 score: 0.937 |
Yong et al. [64] | DGFSD model based on DNN and Graph Convolutional Network (GCN) | Accuracy of 82.78% |
Proposed Method | 8 Classical Machine Learning Model and Stacking-based Meta Classifier | Extra Tree & Stacked Logistic Regression—AUC 0.99 |