Fig. 4

Nomogram development and validation (Adverse foetal/neonatal events). A The nomogram incorporates seven variables, with points allocated according to the scale for each variable. A total score was awarded from the sum of the individual scores, and used to calculate the predicted probability of adverse foetal/neonatal events. B (a, b) Calibration curves for the nomogram in the Delivery (a) and Validation (b) groups. The calibration plot illustrates the accuracy of the original prediction (“Apparent”: light dotted line) and bootstrap models (“Bias-corrected”: solid line) in predicting the probability of adverse foetal/neonatal events. C (a, b) Decision curve analysis for the nomogram in the Delivery (a) and Validation (b) groups. The y-axis indicates the net benefit, which is the sum of the benefits (true positives) minus harm (false positives). The x-axis indicates the threshold probability. The red line represents the net nomogram benefit. The yellow and blue lines represent the hypotheses that all and no patients had adverse foetal/neonatal events, respectively. D (a, b) Violin plot analysis compared the distribution of risk prediction probabilities for those with adverse foetal/neonatal events versus those without in the Delivery (a) and Validation (b) groups. The predicted risk for those with adverse foetal/neonatal events was markedly higher than for those without adverse events in both groups. Demonstration of a violin plot and the depicted data. The three lines within the plot show the 1st and 3rd quartiles and the median of the dataset; the violin body width indicates the density of data along the y-axis. The violin edges represent the minimum and maximum dataset values. AE adverse event, RF respiratory failure