System and method for predicting gestational diabetes mellitus

Abstract

The invention discloses a system and a method for predicting gestational diabetes mellitus, and relates to gestational diabetes mellitus. The system comprises: a data acquisition module for acquiringclinical indexes and clinical diagnosis results; a first index extraction module used for extracting and obtaining associated clinical indexes; a second index extraction module used for randomly combining the associated clinical indexes and calculating a combination with the highest correlation coefficient as an effective clinical index; a training module comprising a logistic regression model, asupport vector machine model, a K nearest neighbor classification model and a deep neural network model which are obtained through training according to effective clinical indexes; and a prediction module used for respectively inputting the effective clinical indexes of a pregnant woman to be predicted into the four models to obtain corresponding prediction results, and predicting the prediction results to obtain the onset risk of the pregnant woman to be predicted. The system and the method have the following beneficial effect that a pregnant woman can conveniently take preventive or interventional measures in advance by predicting the onset risk of gestational diabetes mellitus.

Description

technical field [0001] The invention relates to gestational diabetes, in particular to a system and method for predicting gestational diabetes. Background technique [0002] Gestational diabetes mellitus (GDM) is one of the common complications of pregnancy, with an incidence rate ranging from 1.5% to 25% in various regions of the world, and there are differences in the susceptibility of races. East Asian populations belong to high-risk ethnic groups. The incidence of GDM in Chinese pregnant women can reach about 15%. GDM can lead to macrosomia, premature birth, intrauterine growth restriction, stillbirth, polyhydramnios, neonatal hypoglycemia, neonatal hyperinsulinemia, etc. Adverse events that affect the health and safety of mothers and babies; in the long run, exposure to high-sugar environments in the womb will also increase the risk of non-communicable diseases (such as diabetes, obesity, coronary heart disease, etc.) in offspring as adults. [0003] However, the GDM d...

AI Technical Summary

Problems solved by technology

However, it has not been clinically proven, and these indicators are not routine inspection items in hospitals, so they cannot be widely used in actual clinical work

Images