Predicting Extubation Outcome in Preterm Newborns: a Comparison of Neural Networks with Clinical Expertise and Statistical Modeling
2006

Web-based Tool for Predicting Extubation in Premature Infants

Sample size: 183 publication Evidence: high

Author Information

Author(s): Martina Mueller, Carol L Wagner, David J Annibale, Rebecca G Knapp, Thomas C Hulsey, Jonas S Almeida

Primary Institution: Medical University of South Carolina

Hypothesis

Can a web-based decision-support tool effectively predict extubation outcomes in premature infants with respiratory distress syndrome?

Conclusion

The study developed a web-based decision-support tool that can aid clinicians in making extubation decisions for premature infants, potentially improving patient care.

Supporting Evidence

  • The ANN model showed a significant difference in performance compared to clinical expert predictions.
  • The tool can process user data to provide a prediction score and a novelty index.
  • The study identified 51 potentially predictive variables for extubation decisions.

Takeaway

Doctors can use a special online tool to help decide when to take breathing tubes out of premature babies, making it safer for them.

Methodology

The study used a five-step procedure to identify predictive variables and developed two models (ANN and MLR) to create a decision-support tool.

Potential Biases

Different clinicians may have varying levels of expertise, which could influence the selection of predictive variables.

Limitations

The study was based on retrospective data and may not generalize to all premature infants outside the specified weight range.

Participant Demographics

Infants born between January 1999 and October 2002, with a birth weight between 900 and 1500 grams and diagnosed with RDS.

Statistical Information

P-Value

0.0006

Statistical Significance

p = 0.0006

Digital Object Identifier (DOI)

10.1186/1472-6947-6-11

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication