Application of CEEMD noise reduction algorithm in ultrasound imaging in evaluating fetuses with abnormal glucose metabolism in late pregnancy

Objectives: To explore the predictive effect of abnormal glucose metabolism and fetal hemodynamic parameters on adverse pregnancy outcome.

Methods: One hundred and nine pregnant women with abnormal glucose metabolism during pregnancy from June 2016 to October 2018 were selected and divided into poor prognosis group (34 cases) and good prognosis group (75 cases). The hemodynamic parameters of fetal cerebral artery (MCA), umbilical artery (UA) and uterine artery of pregnancy (UT-A), including peak systolic velocity (s / D), resistance index (RI) and plasticity index (PI), were measured by color Doppler ultrasound. The receiver operating characteristic (ROC) curve of adverse pregnancy outcomes was drawn and the best threshold index was determined.

Results: MCA-PI poor prognosis group, MCA-RI, RI ratio (MCA/UA) are lower than the good prognosis group, Ut-A-PI is higher than the good prognosis group (P<0.05,). ROC curve analysis results show that when the MCA-PI 1.56, the sensitivity of the predicted adverse outcomes of pregnancy, the highest specificity<, was 91.18%, 80.00%, respectively. Logistic regression analysis of risk factors shows poor pregnancy outcomes include: pregnant women, older age, body mass index ≥24.0kg/m2 and a family history of diabetes. Protective factors include exercise during pregnancy, MCA-PI≥1.56, MCA-RI≥0.63 and RI The ratio (MCA/UA) ≥0.84.

Conclusion: Color Doppler ultrasound measured MCA-PI<1.56 the most important indicators of poor pregnancy outcomes as abnormal glucose metabolism during pregnancy and predict the exact cutoff. Pregnant women, older age, body mass index ≥24.0kg/m2 and a family history of diabetes and abnormal glucose metabolism during pregnancy risk factors for adverse outcomes of pregnancy.