The new era of risk assessment for hypertension in pregnancy: From clinical to biochemical markers in a comprehensive predictive model

• DOI: 10.1016/j.tjog.2024.10.014
• Publication Date: 2025-03-04T15:54:22Z
• Authors: Liju Nie; Ziyu Zhang; Qinglan Yao; Huayan Chen; Chao Xu; Lin Chen; Chengcheng Liu; Lantao Tu; Yuping Yi; Tianqiang Huang; Xiaoming Zeng; Lamei Yu

This study aims to develop and validate a model based on the weighted random forest (WRF) algorithm predict early-onset preeclampsia (PE) assess importance of various clinical biochemical markers in early risk identification. was conducted at Jiangxi Maternal Child Health Hospital involved 12,699 pregnant women from January 2019 June 2022. Extensive were collected through prenatal care data, which used construct predictive for PE. The developed using WRF Logistic regression methods, multivariable analysis employed identify significantly associated with relative evaluated (RF) sample 1200 patients diagnosed Blood pressure pre-pregnancy body mass index (BMI) identified as most critical variables affecting accuracy PE prediction model. demonstrated higher (AUC = 0.9614) than 0.9138), highlighting its superiority identification WRF-based this effectively predicts PE, blood BMI vital factors. These findings underscore employing comprehensive assessment pregnancy, facilitating intervention improving health outcomes their newborns.