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

Data mining methods for improving birth outcomes prediction.

Linda K Goodwin1, Mary Ann Iannacchione

  • 1Health Systems and Primary Care, Duke University School of Nursing, USA. linda.goodwin@duke.edu

Outcomes Management
|April 16, 2002
PubMed
Summary

Predicting preterm birth is complex. Data mining identified 7 demographic variables that can predict birth outcomes with 72% accuracy, aiding targeted interventions.

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

Database management systems--their place in nursing informatics education.

Computers, informatics, nursing : CIN·2010
Same author

A topical antioxidant solution containing vitamins C and E stabilized by ferulic acid provides protection for human skin against damage caused by ultraviolet irradiation.

Journal of the American Academy of Dermatology·2008
Same author

Predictors of preterm birth in birth certificate data.

Studies in health technology and informatics·2008
Same author

The vagina dialogues: do you douche?

The American journal of nursing·2004
Same author

Protecting patient privacy in clinical data mining.

Journal of healthcare information management : JHIM·2002

Area of Science:

  • Clinical informatics
  • Biostatistics
  • Public health

Background:

  • Predicting preterm birth is a significant clinical challenge impacting families and healthcare.
  • Existing methods for predicting birth outcomes require improvement for effective intervention.

Purpose of the Study:

  • To explore data mining techniques for predicting preterm birth outcomes.
  • To identify a parsimonious set of predictors for preterm birth.

Main Methods:

  • Exploratory data mining was applied to a racially diverse dataset of 19,970 individuals.
  • Receiver operating characteristic (ROC) analyses were used to evaluate predictor performance.

Main Results:

  • A model using 7 demographic variables achieved an area under the curve (AUC) of 0.72.

Related Experiment Videos

  • This suggests that a limited set of demographic factors can effectively predict preterm birth.

    • Conclusions:

    • Data mining offers a promising approach for predicting preterm birth outcomes.
    • Identifying key demographic predictors can facilitate targeted interventions for improved birth outcomes management.