Related Experiment Video
Updated: Dec 17, 2025
08:58
Development of a Neonatal Piglet Acute Lung Injury Model Recreating the Early Environment of Preterm Infant Lungs
Published on: October 31, 2025
427
Bronchopulmonary dysplasia predicted at birth by artificial intelligence
Henrik Verder1, Christian Heiring2, Rangasamy Ramanathan3
1Department of Pediatrics, Holbaek University Hospital, Holbaek, Denmark.
Acta Paediatrica (Oslo, Norway : 1992)
|June 23, 2020
Summary
A new bedside test accurately predicts bronchopulmonary dysplasia (BPD) in preterm infants at birth. This early detection allows for timely interventions, potentially improving outcomes for infants at risk of BPD.
Area of Science:
- Neonatal Medicine
- Biomarkers
- Artificial Intelligence
Background:
- Bronchopulmonary dysplasia (BPD) is a significant complication in preterm infants.
- Early prediction and intervention are crucial for improving BPD outcomes.
Purpose of the Study:
- To develop a rapid bedside test for predicting BPD in preterm infants.
- To enable early, targeted interventions for BPD.
Main Methods:
- A multicenter study involving preterm infants (24-31 weeks gestational age).
- Combined clinical data at birth with spectral analysis of gastric aspirate samples.
- Utilized artificial intelligence to develop a predictive algorithm for BPD.
Main Results:
- The predictive algorithm achieved 88% sensitivity and 91% specificity for BPD.
- Key predictors included spectral data from gastric aspirates, birth weight, and gestational age.
- Identified critical spectral wave numbers for classification and surfactant treatment.
Conclusions:
- A novel point-of-care test for BPD prediction has been developed.
- The test utilizes a software algorithm combining clinical and spectral data.
- Early BPD diagnosis via this test facilitates targeted interventions and may improve infant outcomes.
