Bronchopulmonary dysplasia predicted at birth by artificial intelligence

Abstract

Aim To develop a fast bedside test for prediction and early targeted intervention of bronchopulmonary dysplasia (BPD) to improve the outcome. Methods In a multicentre study of preterm infants with gestational age 24‐31 weeks, clinical data present at birth were combined with spectral data of gastric aspirate samples taken at birth and analysed using artificial intelligence. The study was designed to develop an algorithm to predict development of BPD. The BPD definition used was the consensus definition of the US National Institutes of Health: Requirement of supplemental oxygen for at least 28 days with subsequent assessment at 36 weeks postmenstrual age. Results Twenty‐six (43%) of the 61 included infants developed BPD. Spectral data analysis of the gastric aspirates identified the most important wave numbers for classification and surfactant treatment, and birth weight and gestational age were the most important predictive clinical data. By combining these data, the resulting algorithm for early diagnosis of BPD had a sensitivity of 88% and a specificity of 91%. Conclusion A point‐of‐care test to predict subsequent development of BPD at birth has been developed using a new software algorithm allowing early targeted intervention of BPD which could improve the outcome.

Corticosteroids in the prevention and treatment of infants with bronchopulmonary dysplasia: Part II. Inhaled corticosteroids alone or in combination with surfactants

“…A recently published multicenter pilot study reported a model using artificial intelligence for the prediction of BPD immediately after birth 56 . An algorithm, based on the analysis of clinical data at birth (surfactant treatment, gestational age, and birth weight) combined with spectral data of gastric aspirate samples taken at birth, could predict BPD with a sensitivity of 88% and a specificity of 91% 56 …”

Early prediction of bronchopulmonary dysplasia is possible and important

“…We would like to thank Fahri Ovali for commenting 1 on our article "Bronchopulmonary dysplasia predicted at birth by artificial intelligence". 2 We agree that optimal early treatments or prophylaxis of bronchopulmonary dysplasia (BPD) has not yet been developed.…”

“…However, there are many potential effective treatment options to be tested. 2 Our screening test, which has a high sensitivity and specificity and is based on a combination of specific prenatal lung biomarkers and clinical data, may enable neonatologists to develop and test treatments for BDP from birth and during the first postnatal days when the immature lung is at its most vulnerable stage. Since BPD is a progressive disease, it is important to identify high-risk patients as early as possible, and any delay in starting preventative treatment could be harmful.…”

Risk Identification of Bronchopulmonary Dysplasia in Premature Infants Based on Machine Learning

“…Ambalavanan et al discovered that risk factors were related to low birth weight, high oxygen demand, male sex, administration of an extra surfactant dose, high oxygen index, and birth status but not related to alveolar oxygen partial pressure (PaO 2 ) and inhaled nitric oxide (iNO) via a decision tree (11). In addition, Verder et al added spectral data to the model, using support vector machine, which predicted BPD with spectral data (12). However, the scale of the data was too small.…”