Study trains AI to successfully detect BPD in premature infants
The rapidly developing area of technology and artificial intelligence (AI) within respiratory medicine and science was under the spotlight at this year’s European Respiratory Society (ERS) Congress, including the use of artificial neural networks (ANNs) in detecting bronchopulmonary dysplasia (BPD) in preterm infants.
ANNs can be trained to detect BPD in preterm infants by analysing their breathing patterns, Swiss researchers reported at the ERS Congress.
There is difficulty in identifying BPD with current lung function tests as they require sophisticated equipment, the study authors said.
ANNs are mathematical models which, once trained with large amounts of data, can be used for classification and prediction.
Lead author Professor Edgar Delgado-Eckert, adjunct professor at the Department of Biomedical Engineering at the University of Basel said: ‘Until recently, this need for large amounts of data has hindered efforts to create accurate models for lung disease in infants because it is so difficult to assess their lung function.’
For this study, the researchers used a simpler and non-invasive alternative: measuring an infant’s inspiratory and expiratory air flow during tidal breathing to yield a large amount of sequential flow data which could be used to train an ANN.
Professor Delgado-Eckert’s team studied a group of 139 term and 190 preterm infants who had been assessed for BPD, recording their breathing using a soft face mask and sensor for 10 minutes while they slept.
Among the 190 preterm infants, 47 were diagnosed with mild BPD, 54 with moderate BPD and 31 with severe BPD.
For each infant, 100 consecutive regular breaths, carefully inspected to exclude sighs or other artefacts, were used to train, validate and test a long short-term memory recurrent ANN.
The data was randomly split into 60% for training and 20% for validation, with the remaining 20% given to the model unseen to test if it could identify infants with BPD.
On the unseen test data, the model achieved 96% accuracy, 100% specificity, 96% sensitivity and 98% precision for detecting BPD.
Professor Delgado-Eckert said: ‘Our research delivers, for the first time, a comprehensive way of analysing the breathing of infants, and allows us to detect which babies have BPD as early as one month of corrected age – the age they would be if they had been born on their due date – by using the ANN to identify abnormalities in their breathing patterns.’
ERS Congress co-chair Professor Judith Löffler-Ragg said the research presented at this year’s event under the theme of ‘Humans and machines: getting the balance right’ was pioneering and should guide future developments.
‘It is extremely important that we view developments in technology, and specifically AI, with an open mind but also a critical eye,’ she said.
‘Our vision is to advance personalised medicine through the responsible use of AI, continuously improving respiratory medicine.’
link