Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

High-risk pregnancy prediction using Taguchi-optimized machine learning methods and TOPSIS-based model selection.

Maryam Mousavi Nogholi1, Ashkan Mozdgir2, Youness Javid1

  • 1Department of Industrial Engineering, Faculty of Engineering, Kharazmi University, Tehran, Iran.

Scientific Reports
|May 26, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effectiveness of silver diamine fluoride in preventing and arresting dental caries in pediatric dentistry.

Dental research journal·2026
Same author

Stem Cell Therapy: A Promising Approach in Treatment of COVID-19.

Current stem cell research & therapy·2020
See all related articles

Machine learning models can predict high-risk pregnancies by analyzing demographic and pregnancy-related factors. This approach aids in the early identification and prioritization of at-risk mothers and fetuses, improving maternal healthcare outcomes.

Area of Science:

  • Maternal and Fetal Health
  • Medical Informatics
  • Machine Learning in Healthcare

Background:

  • High-risk pregnancies present significant maternal and neonatal healthcare challenges.
  • Current clinical tools often fail to identify individual complexities, especially in resource-limited settings.
  • Delayed identification of high-risk pregnancies contributes to elevated mortality rates.

Purpose of the Study:

  • To develop and evaluate machine learning (ML) models for predicting high-risk pregnancies.
  • To identify key predictive features for high-risk pregnancy conditions.
  • To establish an integrated framework for balanced ML model selection in maternal health.

Main Methods:

  • Utilized data from 62 pregnant women in Iran (2014-2016).
Keywords:
Feature selectionMachine learning optimizationMulti-criteria decision analysisPredictive modelingPregnancy risk prediction

Related Experiment Videos

  • Employed feature engineering, Taguchi method for optimization, and five supervised ML models (KNN, RF, DT, SVM, MLP).
  • Integrated feature-group comparison, Taguchi optimization, and TOPSIS for multi-criteria model ranking.

    • Main Results:

    • Pregnancy-related features demonstrated stronger predictive power for high-risk conditions like IUFD, IUGR, and preterm birth.
    • K-Nearest Neighbors (KNN) achieved 88% accuracy and high recall using pregnancy-related features.
    • TOPSIS ranking favored SVM for demographics, KNN for pregnancy-related features, and MLP for the complete case dataset.

    Conclusions:

    • The developed integrated framework supports earlier identification and prioritization of high-risk pregnancies.
    • Machine learning models, particularly KNN with pregnancy-related features, show promise in enhancing prediction accuracy.
    • This approach can assist clinicians and healthcare centers in managing maternal and neonatal risks more effectively.