Fig. 3From: Development and validation of prediction models for gestational diabetes treatment modality using supervised machine learning: a population-based cohort studyPre- and post-calibration plots using logistic regression with level 1–4 predictors on the validation set. The simpler logistic regression model included gestational week at diagnosis of gestational diabetes, the diagnostic fasting glucose value, the status and frequency of self-monitored glycemic control at fasting during 1-week post diagnosis, and an interaction term of the last two variables. The dashed line indicates a perfectly calibrated modelBack to article page