Risk warning model of postoperative adverse pregnancy outcome in patients with cervical incompetence based on decision tree algorithm
10.3969/j.issn.1671-8348.2025.03.019
- VernacularTitle:基于决策树模型建立宫颈机能不全患者术后不良妊娠结局的风险预警模型
- Author:
Jingjing YI
1
;
Xingting LI
;
Chunrong PU
;
Lei CHEN
Author Information
1. 德阳市人民医院产科,四川 德阳 618000
- Keywords:
cervical incompetence;
cervical cerclage;
adverse pregnancy outcome;
influencing factor
- From:
Chongqing Medicine
2025;54(3):668-672,677
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the risk factors of adverse pregnancy outcomes in the patients with cervical incompetence(CI),and to establish a risk warning model of adverse pregnancy outcomes in CI pa-tients based on decision tree model.Methods The clinical data of 159 patients with CI admitted and treated in this hospital from February 2022 to April 2023 were retrospectively analyzed,and the risk factors for adverse pregnancy outcomes were screened and the decision tree model for postoperative adverse pregnancy outcomes was constructed.The internal verification method was 5-fold cross-validation.Results The incidence rate of adverse pregnancy outcome was 22.64%.The pregnant weeks of cervical cerclage,amniotic cystocele,multiple cervical cerclage,preoperative cervical length and amniotic fluid sediment were all influential factors for ad-verse pregnancy outcome occurrence(P＜0.05).The amniotic fluid sediment was the most important factor affecting the postoperative adverse pregnancy outcome in CI patients,and the preoperative cervical length had little influence on the postoperative adverse pregnancy outcomes in CI patients.The area under the curve(AUC)value of logistic regression model was slightly higher than that of the decision tree model.The accura-cy rate of the 5-fold cross-validation model was 78.3%.Conclusion During clinical treatment,the above two models can be combined to find the influencing factors of postoperative adverse pregnancy outcomes in CI pa-tients from different aspects,and provide references for clinical medical staff to evaluate the disease condition of CI patients and formulate the intervention plans.
