From: Early prediction of MODS interventions in the intensive care unit using machine learning
 | MIMIC-III | MIMIC-IV | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Models | AUC | Accuracy | Sensitivity | Specificity | YI | Utility_score | AUC | Accuracy | Sensitivity | Specificity | YI | utility_score |
Super learner | 0.949 ± 0.014 | 0.902 ± 0.013 | 0.896 ± 0.014 | 0.930 ± 0.012 | 0.826 ± 0.011 | 0.780 ± 0.011 | 0.941 ± 0.014 | 0.893 ± 0.011 | 0.884 ± 0.011 | 0.929 ± 0.013 | 0.813 ± 0.014 | 0.763 ± 0.011 |
SubSuperLearner | 0.942 ± 0.012 | 0.901 ± 0.014 | 0.896 ± 0.012 | 0.928 ± 0.014 | 0.824 ± 0.014 | 0.778 ± 0.012 | 0.935 ± 0.012 | 0.888 ± 0.012 | 0.880 ± 0.014 | 0.925 ± 0.014 | 0.809 ± 0.012 | 0.760 ± 0.011 |
DWNN | 0.967 ± 0.012 | 0.891 ± 0.011 | 0.881 ± 0.011 | 0.939 ± 0.013 | 0.820 ± 0.012 | 0.705 ± 0.015 | 0.960 ± 0.013 | 0.882 ± 0.014 | 0.869 ± 0.012 | 0.935 ± 0.012 | 0.804 ± 0.011 | 0.690 ± 0.015 |
lightgbm | 0.964 ± 0.015 | 0.887 ± 0.013 | 0.879 ± 0.012 | 0.927 ± 0.012 | 0.806 ± 0.012 | 0.759 ± 0.013 | 0.959 ± 0.014 | 0.884 ± 0.014 | 0.873 ± 0.012 | 0.930± 0.015 | 0.903 ± 0.012 | 0.738 ± 0.015 |
random forest | 0.963 ± 0.014 | 0.886 ± 0.012 | 0.876 ± 0.012 | 0.932 ± 0.011 | 0.808 ± 0.015 | 0.734 ± 0.012 | 0.958 ± 0.012 | 0.878 ± 0.014 | 0.863 ± 0.015 | 0.943 ± 0.012 | 0.806 ± 0.012 | 0.711 ± 0.012 |
XGBoost | 0.959 ± 0.014 | 0.887 ± 0.012 | 0.882 ± 0.012 | 0.910 ± 0.012 | 0.792 ± 0.013 | 0.756 ± 0.014 | 0.953 ± 0.012 | 0.878 ± 0.014 | 0.868 ± 0.013 | 0.921 ± 0.011 | 0.789 ± 0.012 | 0.730 ± 0.012 |
AdaBoosting | 0.958 ± 0.014 | 0.873 ± 0.011 | 0.868 ± 0.012 | 0.897 ± 0.012 | 0.765 ± 0.012 | 0.726 ± 0.011 | 0.954 ± 0.013 | 0.867 ± 0.013 | 0.853 ± 0.012 | 0.928 ± 0.013 | 0.781 ± 0.012 | 0.705 ± 0.013 |
Logistic Regression | 0.955 ± 0.014 | 0.883 ± 0.014 | 0.875 ± 0.012 | 0.919 ± 0.015 | 0.794 ± 0.015 | 0.678 ± 0.013 | 0.944 ± 0.013 | 0.872 ± 0.014 | 0.862 ± 0.015 | 0.919 ± 0.013 | 0.781 ± 0.012 | 0.657 ± 0.015 |
Naïve Bayes | 0.938 ± 0.012 | 0.834 ± 0.013 | 0.815 ± 0.015 | 0.923 ± 0.011 | 0.738 ± 0.012 | 0.657 ± 0.014 | 0.936 ± 0.013 | 0.810 ± 0.014 | 0.779 ± 0.012 | 0.942 ± 0.012 | 0.721 ± 0.015 | 0.620 ± 0.012 |
KNN | 0.938 ± 0.013 | 0.850 ± 0.013 | 0.833 ± 0.014 | 0.932 ± 0.012 | 0.765 ± 0.012 | 0.616 ± 0.011 | 0.920 ± 0.011 | 0.828 ± 0.012 | 0.804 ± 0.012 | 0.932 ± 0.015 | 0.736 ± 0.014 | 0.608 ± 0.013 |
Decision Tree | 0.928 ± 0.012 | 0.852 ± 0.014 | 0.839 ± 0.011 | 0.912 ± 0.012 | 0.751 ± 0.015 | 0.667 ± 0.014 | 0.914 ± 0.011 | 0.823 ± 0.012 | 0.798 ± 0.015 | 0.926 ± 0.014 | 0.724 ± 0.011 | 0.610 ± 0.012 |