Fig. 3From: Estimation of postpartum depression risk from electronic health records using machine learningPerformance of PPD prediction models with geographical split (a) and temporal split (b) of the training/testing sets. The models are compared by the area under the ROC curve (AUC) and by the sensitivity at false positive rate of 0.2. EHR: risk score based on Electronic Health Records; EPDS: score based on Edinburgh questionnaireBack to article page