Fig. 2: Classification and regression tree (CART) model for predicting severe BPD at 7 and 14 days of age. | Pediatric Research

Fig. 2: Classification and regression tree (CART) model for predicting severe BPD at 7 and 14 days of age.

From: Development and validation of a prediction model for bronchopulmonary dysplasia using respiratory severity score

Fig. 2: Classification and regression tree (CART) model for predicting severe BPD at 7 and 14 days of age.

a Postnatal day 7 model. b Postnatal day 14 model. Predictive models with and without severe BPD were created using classification and regression tree (CART) analysis. The cutoff value for each root node or daughter node is automatically selected according to the likelihood-ratio chi-square statistic. The following covariates were used to create the prediction model, RS score, gestational age, birth weight, clinical CAM, histologic CAM, PROM, 1 min Apgar score, 5 min Apgar score, RDS, symptomatic PDA, IVH, Sepsis, and intubation at birth. The percentage represents predicted probabilities, while the N value represents the number of infants in each category. BPD is bronchopulmonary dysplasia, CAM is chorioamnionitis, PROM is premature rupture of membrane, RDS is respiratory distress syndrome, PDA is patent ductus arteriosus, IVH is intraventricular hemorrhage, and RS score is respiratory severity score.

Back to article page