Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
November 1, 2022Annals of Translational MedicineOpen Access

Development and validation of the prediction models for preeclampsia: a retrospective, single-center, case-control study

View Full Paper
Ask AI
Bookmark
Share

Key result

Random forest model outperforms logistic regression for predicting preeclampsia with an AUC of ~0.87.

  • P<0.05
  • n=916

Why the study?

Accurate screening methods for preeclampsia are a current clinical focus due to its major impact on maternal and infant outcomes.

Does a random forest model accurately predict preeclampsia in pregnant women compared to traditional models?

Population

916 pregnant women at the Second Hospital of Tianjin Medical University

Comparison

Random forest model vs logistic regression and classification tree models

Design

Retrospective single-center case-control study

Authors

XCXuhong ChenYLYuan LiJZJi Zhen

Discussion

Loading...

Member takes

Overview

May enhance preeclampsia risk prediction; leaves open external validation before clinical adoption.

Study Design

Type

Case-Control (n=916)

Multicenter

No

Structured PICO

Does a random forest model accurately predict preeclampsia in pregnant women compared to traditional models?

P
Population
916 pregnant women (237 with preeclampsia) evaluated in a retrospective case-control study to develop and validate prediction models.
E
Exposure
Random forest (RF) prediction model for preeclampsia.
C
Comparator
Logistic regression (LR) and classification tree (CT) prediction models.
O
Outcome
Predictive performance for preeclampsia assessed by area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).surrogate

Main Result

p-value: p=<0.05

A random forest machine learning model demonstrated superior predictive performance for preeclampsia compared to logistic regression and classification tree models.

Cite This Study

Chen et al. (2022) conducted a case-control in Preeclampsia (n=916). Random forest (RF) prediction model vs. Logistic regression (LR) and classification tree (CT) models was evaluated on Area under the curve (AUC) for predicting preeclampsia (p=<0.05). A random forest model for predicting preeclampsia achieved an AUC of 0.871, outperforming logistic regression (AUC 0.778) and classification tree (AUC 0.850) models (P<0.05 for all comparisons).

synapsesocial.com/papers/6ab1fc7fc9fd9fb93fff3f37https://doi.org/10.21037/atm-22-4192
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1ACOG Practice Bulletin No. 203: Chronic Hypertension in Pregnancy2019 · 868 citations
  2. 2Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies2004 · 12,885 citations
  3. 3The 2011 Survey on Hypertensive Disorders of Pregnancy (HDP) in China: Prevalence, Risk Factors, Complications, Pregnancy and Perinatal Outcomes2014 · 225 citations
  4. 4Imbalances in circulating angiogenic factors in the pathophysiology of preeclampsia and related disorders2020 · 329 citations
  5. 5Improving preeclampsia risk prediction by modeling pregnancy trajectories from routinely collected electronic medical record data2022 · 78 citations