Disease's type
GDM
Experimental grouping
Modeling training set(4200 cases,GDM(n=1225),Normal Glucose Tolerance(n=2975) and a validation set (1800 cases,GDM(n=519),Normal Glucose Tolerance(n=1281)
GPT's summary
This study developed and validated an early pregnancy risk prediction model for gestational diabetes mellitus (GDM) based on a cohort of 6000 pregnant women. Independent risk factors identified for GDM included age, pre-pregnancy BMI, glycosylated hemoglobin (HbA1c), blood uric acid (UA), triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C). The model demonstrated good predictive performance, with an AUC of 0.803 in the training cohort and 0.782 in the validation cohort, indicating reliable sensitivity and specificity. These findings suggest that combining maternal demographic and laboratory indicators in early pregnancy can help identify high-risk women, providing a basis for early screening and intervention.
RF's name
High Density Lipoprotein Cholesterol
Sample's type
Serum
Gestational weeks
8th to 12th gestational weeks
Experiemental methods
Enzymatic Methods
Machine learning algorithms
Logistic Regression
Title
Establishment of gestational diabetes risk prediction model and clinical verification
Evidence's type
Risk factor
Year
2024
Journal
Journal of Endocrinological Investigation
PMID