Table 1.
Patient characteristics and hospital characteristics (N = 22,926).
Table 2.
The LR model using selected variables related to in-hospital mortality.
Figure 1.
Schematic representation of artificial neural network model with 6 input nodes, 3 nodes in a single hidden layer, and a single output node representing in-hospital mortality.
X1, age; X2, gender; X3, Charlson co-morbidity index; X4, hospital volume; X5, surgeon volume; X6, length of stay; IB, input layer bias; HB, hidden layer bias.
Table 3.
Comparison of 1000 pairs of ANN and LR models for predicting in-hospital mortality.
Table 4.
Global sensitivity analysis of the ANN model in predicting in-hospital mortality.
Table 5.
Comparative performance indices of ANN and LR models when using 100 new data sets to predict in-hospital mortality.