Related Experiment Video
Updated: Jun 30, 2025
Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
Deep Learning Model for Prediction of Bronchopulmonary Dysplasia in Preterm Infants Using Chest Radiographs
Hao-Yang Chou1, Yung-Chieh Lin2, Sun-Yuan Hsieh1,3,4,5,6,7
1Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan, 70101, Taiwan.
Insights
Artificial intelligence (AI) accurately diagnoses bronchopulmonary dysplasia (BPD) in preterm infants using chest radiographs. This AI model offers early detection, surpassing expert diagnostic accuracy for improved infant lung health.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Pediatrics
- Neonatal Respiratory Medicine
Background:
- Bronchopulmonary dysplasia (BPD) is a prevalent complication in preterm infants, often leading to pulmonary vascular disease and impaired lung function.
- Accurate and timely diagnosis of BPD is crucial for effective management and improved outcomes in premature neonates.
Purpose of the Study:
- To develop and evaluate an artificial intelligence (AI) model for the accurate and efficient diagnosis of BPD in preterm infants.
- To enable early detection of BPD using chest radiographs within 24 hours of birth.
Main Methods:
- Retrospective analysis of two datasets: 1491 chest radiographs for lung segmentation and 1021 for BPD prediction in preterm infants.
- Application of transfer learning for lung region segmentation and image fusion techniques to enhance AI model performance.
- Evaluation of the AI model's diagnostic performance against expert clinicians and established BPD criteria (NICHD and Jensen).
Main Results:
- The AI lung segmentation model achieved a high dice score of 0.960 for preterm infants.
- The BPD prediction model demonstrated superior diagnostic performance compared to human experts.
- Consistent AI model performance was observed for radiographs taken within 24 hours and those taken between 25 to 168 hours postnatal age.
Conclusions:
- This study introduces the first deep learning model for BPD prediction using preterm chest radiographs, enabling detection in under 24 hours.
- The AI model significantly surpasses expert diagnostic accuracy in predicting lung development and identifying BPD in preterm infants.
- AI-driven analysis holds promise for timely and accurate BPD diagnosis, potentially improving clinical management and long-term outcomes.
Abstract:
Bronchopulmonary dysplasia (BPD) is common in preterm infants and may result in pulmonary vascular disease, compromising lung function. This study aimed to employ artificial intelligence (AI) techniques to help physicians accurately diagnose BPD in preterm infants in a timely and efficient manner. This retrospective study involves two datasets: a lung region segmentation dataset comprising 1491 chest radiographs of infants, and a BPD prediction dataset comprising 1021 chest radiographs of preterm infants. Transfer learning of a pre-trained machine learning model was employed for lung region segmentation and image fusion for BPD prediction to enhance the performance of the AI model. The lung segmentation model uses transfer learning to achieve a dice score of 0.960 for preterm infants with 168 h postnatal age. The BPD prediction model exhibited superior diagnostic performance compared to that of experts and demonstrated consistent performance for chest radiographs obtained at 24 h postnatal age, and those obtained at 25 to 168 h postnatal age. This study is the first to use deep learning on preterm chest radiographs for lung segmentation to develop a BPD prediction model with an early detection time of less than 24 h. Additionally, this study compared the model's performance according to both NICHD and Jensen criteria for BPD. Results demonstrate that the AI model surpasses the diagnostic accuracy of experts in predicting lung development in preterm infants.
