Skip to main content

Table 2 Performance of various models for COPD diagnosis in test set

From: Development and application of a deep learning-based comprehensive early diagnostic model for chronic obstructive pulmonary disease

Models(n)

AUC

Accuracy

Precision

Recall

F1-score

Brier score

DL (n = 2,983)

0.844

0.782(0.747,0.782)

0.822(0.822,0.829)

0.782(0.747,0.782)

0.792(0.771,0.792)

0.154

RA (n = 2,983)

0.944

0.882(0.882,0.908)

0.882(0.882,0.906)

0.882(0.882,0.908)

0.882(0.882,0.907)

0.085

DL + RA (n = 2,983)

0.952

0.869(0.869,0.886)

0.887(0.887,0.909)

0.869(0.869,0.886)

0.873(0.873,0.893)

0.091

DL + RA + EF (n = 2,317)

0.971

0.886(0.856,0.886)

0.909(0.900,0.909)

0.886(0.856,0.886)

0.891(0.867,0.891)

0.080

  1. RA: Radiomic Feature, DL: Deep Learning Feature, EF: Epidemiological Feature