Zheng T, et al. - Researchers sought to construct a simple model using biochemical markers and machine learning algorithm to prognosticate gestational diabetes mellitus (GDM) in early pregnancy in Chinese women. The GDM risk-prediction model was built using data from 4771 pregnant women in early gestation. The prediction model addressed the correlation of predictors and incorporated maternal age, prepregnancy body mass index, fasting plasma glucose and triglyceride and had a predictive accuracy of 0.64 and an area under receiver operating characteristic curve of 0.766 (95% CI 0.731, 0.801). The simplicity, as well as the potential applicability of this GDM prediction model in Chinese women, was suggested.