Using Vital Signs for the Early Prediction of Necrotizing Enterocolitis in Preterm Neonates with Machine Learning
2024

Using Vital Signs to Predict Necrotizing Enterocolitis in Preterm Infants

Sample size: 267 publication Evidence: moderate

Author Information

Author(s): Rosa Verhoeven, Thijmen Kupers, Celina L. Brunsch, Jan B. F. Hulscher, Elisabeth M. W. Kooi

Primary Institution: University Medical Center Groningen, University of Groningen

Hypothesis

Can vital signs measured in the first days after birth predict necrotizing enterocolitis (NEC) in preterm infants?

Conclusion

Vital signs can predict NEC in preterm infants within the first 5 postnatal days, with splanchnic and cerebral oxygenation being the most significant predictors.

Supporting Evidence

  • 32 out of 267 patients developed NEC.
  • Support vector machine and logistic regression models achieved an F1-score of 0.82.
  • Splanchnic and cerebral oxygenation contributed 40.1% and 24.8% to the prediction, respectively.

Takeaway

Doctors can use early vital signs to tell if a premature baby might get a serious tummy problem called NEC, helping them act quickly to prevent it.

Methodology

Machine learning algorithms were trained on vital signs data from preterm infants to predict NEC.

Potential Biases

The retrospective design may introduce selection bias and the imputation of missing data could affect the reliability of predictions.

Limitations

The study faced challenges with missing data, particularly in splanchnic and cerebral oxygenation measurements.

Participant Demographics

267 preterm infants with gestational age < 30 weeks, including 32 who developed NEC.

Statistical Information

P-Value

0.001

Statistical Significance

p < 0.001

Digital Object Identifier (DOI)

10.3390/children11121452

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication