Mode
Text Size
Log in / Sign up

Can breathing patterns detected by AI help diagnose bronchopulmonary dysplasia in infants?

moderate confidence  ·  Last reviewed May 19, 2026

Bronchopulmonary dysplasia (BPD) is a chronic lung disease that mainly affects premature infants. Diagnosing BPD traditionally requires sophisticated equipment or active patient cooperation, which is difficult in infants. Researchers are exploring whether artificial intelligence (AI) can analyze simple breathing patterns to detect BPD. Early evidence suggests AI can identify BPD from tidal breathing (normal, quiet breathing) with high accuracy, potentially offering a non-invasive, low-cost screening tool.

What the research says

A 2024 study tested a type of AI called a long short-term memory (LSTM) network on tidal breathing flow time series from 329 infants. The model detected BPD with 97% accuracy, showing that subtle breathing pattern changes contain clinically useful information 1. This approach is promising because tidal breathing measurements are non-invasive and require no patient cooperation, unlike conventional lung function tests 1.

Other research supports the value of analyzing continuous vital signs in preterm infants. A review notes that subtle changes in heart rate, respiratory rate, and oxygen saturation patterns can predict BPD, among other conditions, and that AI analysis can detect patterns missed by clinicians 5. Lung ultrasound scores have also been linked to BPD severity, with higher scores in moderate-to-severe cases, though many infants with BPD still had normal ultrasounds 4.

While AI breathing analysis is not yet standard practice, it fits into a broader trend of using big-data analytics to improve neonatal care 5. The LSTM model's high accuracy suggests it could become a useful screening tool, but more studies are needed to confirm its reliability across different populations and settings.

What to ask your doctor

  • Is AI-based breathing pattern analysis available at our hospital for diagnosing BPD?
  • How does the accuracy of AI analysis compare with traditional diagnostic methods for BPD?
  • Could this technology reduce the need for invasive or complex lung function tests in my baby?
  • Are there any ongoing studies or clinical trials using AI for BPD diagnosis that my baby could join?
  • What are the current standard methods for diagnosing BPD, and how might AI change that in the future?

This question is drawn from common patient questions about this topic and answered using cited medical research. We do not provide individualized advice.