Skip to main content

Table 2 Performances of the DIGIROP-Birth model for TR-ROP in the study cohort (n = 442)

From: Validation of the DIGIROP-birth model in a Chinese cohort

Cutoff

0.0084 probability

0.0076 probability

0.0083 probability

0.0089 probability

0.0200 probability

Sensitivity

48/93 (51.6%)

49/93 (52.7%)

48/93 (51.6%)

46/93 (49.5%)

32/93 (34.4%)

Specificity

263/349 (75.4%)

257/349 (73.6%)

263/349 (75.4%)

263/349 (75.4%)

311/349 (89.1%)

PPV

48/134 (35.8%)

49/141 (34.8%)

48/134 (35.8%)

46/132 (34.8%)

32/70 (45.7%)

NPV

263/308 (85.4%)

257/301 (85.4%)

263/308 (85.4%)

263/310 (84.8%)

311/372 (83.6%)

  1. TR-ROP Treatment-requiring retinopathy of prematurity, PPV Positive predictive value, NPV Negative predictive value