Skip to main content

Table 2 Performance metrics for different prediction models*

From: Machine learning in a real-world PFO study: analysis of data from multi-centers in China

 

C-index

95%CI

Brier Score

95%CI

Cox model

0.54

0.20–0.97

0.024

0.007–0.048

RFS model

0.87

0.71–0.98

0.026

0.011–0.046

  1. *Both models were constructed from the top ranking features, including fasting blood glucose, thickness of interventricular septum, the ratio of mitral peak early (E) to late (A) diastolic filling velocity, left ventricular end-systolic dimension, body mass index, systolic blood pressure, thickness of the posterior wall and PTA
  2. RFS random forest survival, CI confidence incidence