Fig. 1From: Fetal weight estimation based on deep neural network: a retrospective observational studyThe flow chart of cohort selection, feature distribution and Deep Neural Network (DNN) model. A Cohort selection. Pregnancies were first identified by offspring birth date. Cases with anomalous fetuses, multifetal gestations and without Ultrasound examination within 2 weeks before delivery were excluded. Finally, the cohort was divided into training and validation sets (see Methods). B DNN model. C Feature availability distribution. Pie charts are divided according to the sum of features in each feature setBack to article page