Construction of Prediction Model for Preterm Birth in Early Pregnancy via Cervical Elastography
10.3969/j.issn.1005-5185.2023.12.012
- VernacularTitle:基于宫颈弹性成像构建早孕期早产预测模型
- Author:
Sihan CHEN
1
;
Bing HU
;
Xiaozhen XIANG
;
Ling DING
;
Ying YANG
Author Information
1. 三峡大学附属仁和医院超声科,湖北 宜昌 443001
- Keywords:
Premature birth;
Ultrasonography,Doppler,color;
Elasticity imaging techniques;
Cervix uteri;
Prediction model
- From:
Chinese Journal of Medical Imaging
2023;31(12):1298-1303
- CountryChina
- Language:Chinese
-
Abstract:
Purpose To preliminarily establish a predictive model for assessing preterm birth in the first trimester via clinical history and multiple ultrasound parameters.Materials and Methods This study included 200 women in the first trimester of pregnancy from 6 to 8 weeks in the Affiliated Renhe Hospital of Three Gorges University from September 2020 to September 2021,and their clinical history,two-dimensional imaging data and cervical elastography data were collected.Logistic regression analysis,screening and assignment were performed to initially establish a clinical prediction model for evaluating preterm birth during the first trimester.Results Finally,41 pregnant women developed preterm birth(preterm birth group),and 159 pregnant women did not develop preterm birth(term group).History of miscarriage,history of preterm birth,uterocervical angle(UCA),A,and A1 were the independent risk factors for preterm birth via univariate analysis,and multivariate analysis was carried out to obtain the formula:Logit(P)=1.495×abortion history+1.060×A1+0.795×UCA+1.354×A-14.951,which had a good fit via Hosmer-Lemeshow goodness-of-fit test,and the diagnostic efficiency was 96.9%.Conclusion The early pregnancy prediction model can effectively predict the occurrence of preterm birth in pregnant women with singleton pregnancy.
