Skip to main content

Table 5 Diagnostic performance of LightGBM model for SMPP in Prospective validation cohort

From: Development and validation of an early diagnosis model for severe mycoplasma pneumonia in children based on interpretable machine learning

Model

AUC

Precision

F1 score

Sensitivity

Specificity

Negative

Kappa

  

prediction rate

 

LightGBM

91.67%

84.21%

86.49%

88.89%

92.86%

95.12%

0.8047

  1. AUC: Receiver operating characteristic curve