Skip to main content

Table 4 SOFA score prediction

From: Data analytics and clinical feature ranking of medical records of patients with sepsis

method R2 RMSE MAE MSE SMAPE
DL 0.73±0.05 1.85±0.26 1.31±0.12 3.33±0.99 0.34±0.04
SVM (linear) 0.78±0.06 1.81±0.18 1.34±0.13 3.27±0.60 0.34±0.05
MLP 0.76±0.06 1.82±0.13 1.36±0.11 3.29±0.71 0.34±0.04
SVM (kernel) 0.74±0.06 1.82±0.25 1.35±0.13 3.37±1.06 0.33±0.03
RF 0.72±0.04 1.83±0.26 1.32±0.15 3.41±1.01 0.35±0.03
DT 0.48±0.11 2.46±0.28 1.80±0.18 6.11±1.38 0.49±0.07
k-NN 0.41±0.10 2.65±0.35 1.95±0.22 7.12±1.85 0.53±0.06
  1. Performance of the learned models with the different methods evaluated with the different metrics, expressed in the format “average value ± standard deviation”, obtained on 100 executions. DT: decision tree. k-NN: k-nearest neighbors. DL: deep neural network with 3 hidden layers and weight decay. MLP: multi-layer perceptron neural network. RF: random forest. SVM (kernel): support vector machine with kernel. SVM (linear): linear support vector machine. RMSE: root mean square error. MAE: mean absolute error. MSE: mean square error. SMAPE: symmetric mean absolute percentage error. R2: coefficient of determination. RMSE, MAE, MSE, SMAPE: best value 0.00 and worst value +. R2: best value 1.00 and worst value −
\