Early hospital mortality prediction using vital signals

18 Mar 2018 Reza Sadeghi Tanvi Banerjee William Romine

Early hospital mortality prediction is critical as intensivists strive to make efficient medical decisions about the severely ill patients staying in intensive care units. As a result, various methods have been developed to address this problem based on clinical records... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Mortality Prediction MIMIC-III Random Forest F1 score 0.97 # 1
Precision 0.97 # 1
Recall 0.97 # 1
Mortality Prediction MIMIC-III Gaussian SVM F1 score 0.96 # 2
Precision 0.95 # 2
Recall 0.96 # 2
Mortality Prediction MIMIC-III Decision Tree F1 score 0.91 # 3
Precision 0.90 # 4
Recall 0.92 # 3
Mortality Prediction MIMIC-III K-NN F1 score 0.82 # 5
Precision 0.80 # 5
Recall 0.85 # 4
Mortality Prediction MIMIC-III Logistic regression F1 score 0.72 # 6
Precision 0.77 # 7
Recall 0.67 # 6
Mortality Prediction MIMIC-III Linear Discriminant F1 score 0.71 # 7
Precision 0.78 # 6
Recall 0.66 # 7
Mortality Prediction MIMIC-III Linear SVM F1 score 0.70 # 8
Precision 0.80 # 5
Recall 0.63 # 8
Mortality Prediction MIMIC-III Boosted Trees F1 score 0.87 # 4
Precision 0.91 # 3
Recall 0.83 # 5

Methods used in the Paper