Skip to main content

Table 4 Performance of the three models in the validation set

From: Harness machine learning for multiple prognoses prediction in sepsis patients: evidence from the MIMIC-IV database

Models

Precision (95%CI)

Accuracy (95%CI)

Recall (95%CI)

F1-score (95%CI)

AUC (95%CI)

LR

0.672(0.655, 0.689)

0.619(0.604, 0.634)

0.619(0.603, 0.634)

0.636(0.621, 0.652)

0.747(0.732, 0.760)

RF

0.666(0.647, 0.686)

0.714(0.700, 0.728)

0.714(0.700, 0.730)

0.664(0.647, 0.682)

0.755(0.742, 0.769)

CatBoost

0.684(0.669, 0.701)

0.652(0.637, 0.666)

0.652(0.637, 0.666)

0.665(0.650, 0.679)

0.771(0.758, 0.785)

  1. Note: Due to the unbalanced classification of this dataset, we adopted a weighted calculation method for the above multiple classification indexes and assigned different weights to each group
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us