Research Article
Machine Learning-Based Model to Predict the Disease Severity and Outcome in COVID-19 Patients
Table 1
Related studies on mortality prediction for COVID-19 patients.
| Reference | Technique | Dataset | Target class | Result |
| [10] | XGB | 404 patients | Death, survived | 0.95 precision | 0.90 accuracy |
| [11] | XGB | 1747 COVID-19 patients | Fatal, severe | Accuracy 0.668 (fatality) | 0.712 (severe) |
| [12] | SVM | 336 COVID-19 patients | Severe, critical | 0.775 accuracy | [13] | SVM | 137 COVID-19 patients | Severe, nonsevere | 0.815 accuracy |
| [14] | LR | 115 COVID-19 patients | Severe, nonsevere | 0.881 AUROC | 0.839 sensitivity | 0.794 specificity |
| [5] | SVM | 303 patients | Negative, positive cases | 0.967 accuracy |
| [15] | XGB | 3927 COVID-19 patients | ā | 0.85 accuracy | 0.90 AUC |
| [16] | LR | 1696 COVID-19 patients | Home, deceased | 0.89 AUC | 0.82 sensitivity | 0.81 specificity |
| [17] | SVM (linear) | 8000 COVID-19 patients | Mortality, recovered | 0.962 AUC | 0.92 sensitivity | 0.91 specificity |
|
|