Method for establishing model for early prediction of gestational diabetes mellitus

Abstract

The invention discloses a method for establishing a model for early prediction of gestational diabetes mellitus, comprising the following steps: 1, clinically retrospectively collecting a plurality ofpregnant women with normal blood glucose during pregnancy and gestational diabetes mellitus, taking the pregnant women as total collection cases, and collecting information data of each case; 2, randomly extracting a plurality of cases from the total collection cases in the step 1 to serve as a training set; compiling a LightGBM model in the form of 'yes-no' by using a Python language and spyderor jupyter notebook software; setting model parameters, then importing information data of each case of the training set and a result of finally judging whether the patient suffers from gestational diabetes mellitus or not, and constructing a LightGBM prediction model; and 3, inputting the information data of the pregnant woman to be predicted into the lightBGM prediction model constructed in thestep 2 to obtain a prediction risk value of the gestational diabetes mellitus risk in the future, and determining whether the pregnant woman will be sick or not according to the prediction risk value.The method has the advantages of simple and convenient implementation, high result accuracy and the like.

Description

technical field [0001] The present invention relates to a prediction method, in particular to a method for early prediction of gestational diabetes. Background technique [0002] Gestational diabetes mellitus (GDM) refers to abnormal glucose metabolism that occurs for the first time during pregnancy, and all patients have no history of diabetes before pregnancy. With the prevalence of high-calorie diets, inactive lifestyles, and increasing gestational age, the incidence of GDM is increasing year by year. According to the map data released by the International Diabetes Federation (IDF) in 2019, 20 million (1 / 6) pregnant women worldwide have some form of gestational hyperglycemia, of which 84% are gestational diabetes mellitus (GDM). IDF estimates that before 2030, 18.3 million newborns will be affected by gestational high glucose. Multi-center data in my country show that the incidence of GDM is 17.5%. With the opening of the second-child policy in my country, this proportio...

AI Technical Summary

Problems solved by technology

[0003] However, there is currently a lack of simple, intuitive and economical early prediction methods for GDM in clinical practice.

The latest guideline (2019ADA) recommends that pregnant women generally undergo a 75g-glucose tolerance test (OGTT) at 24-28 weeks of pregnancy to diagnose GDM, but the guideline only lists risk factors, lacks quantifiable evaluation criteria, and cannot be individualized The risk of GDM in pregnant women in the second and third trimester

Images