Machine Learning Predictive System for Pre-eclampsia Detection Shows High Accuracy

Strategic Insights -

A team of researchers at Chi Mei Medical Center in Tainan, Taiwan, has developed a machine learning (ML) predictive model that demonstrates high accuracy in detecting pre-eclampsia in pregnant women. According to the study, which was published in the BMJ Health & Care Informatics, the model uses routinely collected clinical data, including gestational age, body weight, blood pressure, and medical history, to predict the risk of pre-eclampsia. The researchers trained five ML models, including logistic regression, random forest, and extreme gradient boosting (XGBoost), and found that the XGBoost model showed the best performance with an accuracy of 0.921. The study's findings suggest that ML-based predictive systems, such as the one developed in this study, could be used to support early intervention and reduce the risk of pre-eclampsia complications.

Key Takeaways:

  • The researchers analyzed data from 2444 pregnant women who delivered at Chi Mei Medical Center between 2015 and 2019.
  • The study found that pre-eclampsia was defined as blood pressure 140/90 mm Hg with proteinuria.
  • The top predictors of pre-eclampsia identified by the SHAP analysis were diastolic blood pressure, systolic blood pressure, and urine glucose.
  • The XGBoost model showed the best performance with the highest accuracy, sensitivity, specificity, and area under the receiver operating characteristics curve of 0.921.
  • The study's findings demonstrate that ML, particularly XGBoost, can effectively predict the risk of pre-eclampsia using standard clinical data.
  • The XGBoost model offers a cost-effective and accurate method for pre-eclampsia risk prediction, enabling real-time assessment and supporting early intervention.

Statistics:

  • 2444 pregnant women were included in the study.
  • 70% of the data was used for training, and 30% for validation.
  • The accuracy of the XGBoost model was 0.921.
  • The sensitivity, specificity, and area under the receiver operating characteristics curve of the XGBoost model were also reported.
  • The researchers plan to focus on larger data sets and clinical integration in future studies.

Sources:

  • NewsRx. Findings from Chi Mei Medical Center Has Provided New Information about Eclampsia (Machine learning predictive system to predict the risk of developing pre-eclampsia). OBGYN & Reproduction Week. October 27, 2025; p 168.
  • Liu, C-F., et al. "Machine learning predictive system to predict the risk of developing pre-eclampsia." BMJ Health & Care Informatics, 2025;32(1).