Additional file 1 of Early prediction of need for invasive mechanical ventilation in the neonatal intensive care unit using artificial intelligence and electronic health records: a clinical study

Additional file 1: Table 1. The percentage of missing data. Table 2. The proposed model architecture. hidden_dim=32, input_dim=83. Table 3. Significance testing of proposed model and existing models. Fig. 1. Sequence windowing. Yellow box means current EHR data and blue box means sequence of recodes to support model training and prediction when sequence length is L. Red box is target label according to sequence window. Fig. 2. Sub-group analysis. (a) Area under the curve (AUC) according to gestational age and (b) AUC according to birth weight.