Radiomics based of deep medullary veins on susceptibility-weighted imaging in infants: predicting the severity of brain injury of neonates with perinatal asphyxia

Table 3 Accuracy and predictive value between three models

	AUC	Accuracy	95% CI	Sensitivity	Specificity	PPV	NPV
Training cohort
Clinical model	0.63	0.754	0.671–0.824	0.791	0.526	0.910	0.294
Radiomics model	0.84	0.784	0.704–0.850	0.784	0.783	0.690	0.855
Combined model	0.84	0.761	0.680–0.831	0.824	0.723	0.646	0.870
Validation cohort
Clinical model	0.59	0.750	0.616–0.856	0.784	0.400	0.930	0.154
Radiomics model	0.80	0.678	0.540–0.797	0.620	0.714	0.565	0.758
Combined model	0.79	0.679	0.540–0.797	0.714	0.657	0.556	0.793

ISSN: 2047-783X