Study trains AI to successfully detect BPD in premature infants

8th October 2024

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.’

Diagnostic aid quickly and reliably identifies respiratory pathogens in critically ill children

16th January 2023

A diagnostic aid based on polymerase chain reactions (PCR) that uses a 52-pathogen custom array card, has been found to provide both rapid (compared to blood culture) and reliable information on respiratory infections in critically ill, mechanically ventilated children, according to a study by UK researchers.

Respiratory tract infections are responsible for a large number of admissions to paediatric intensive care units. Moreover, an intensive care unit is unique environment and for which clinicians often make decisions to use antibiotics with some degree of diagnostic uncertainty.

This was clearly illustrated in one study of paediatric intensive care unit children, where despite most critically children receiving antimicrobial therapy, infection was often not microbiologically confirmed.

While in many cases respiratory infections are viral in nature, it is necessary to utilise methods such as quantitative PCR, as a diagnostic aid to identify the presenting pathogens.

In fact, a recent study in adults found that multiplex bacterial PCR examination of bronchoalveolar lavage, reduced the duration of inappropriate antibiotic therapy of patients admitted to hospital with pneumonia and who were at risk of Gram-negative infection.

In the current study, researchers made use of the TaqMan Array Card (TAC) as a diagnostic aid, which is a microfluidic quantitative PCR system comprising of 384 wells containing pre-aliquoted customised primer and probe combinations.

The aid has been previously shown to be of value in supporting ventilator-associated pneumonia (VAP) diagnosis in adults. Nevertheless, it has not been examined in critically ill children and therefore, the aim of the present study was to assess the utility of TAC to identify bacterial and fungal respiratory pathogens in critically ill children with suspected community acquired pneumonia or VAP.

The study recruited children ≤ 18 years of age who were mechanically ventilated and had commenced or were commencing antimicrobial therapy for a lower respiratory tract infection.

The researchers determined the sensitivity and specificity of TAC to detect bacterial and fungal pathogens causing lower respiratory tract infections and the time to a result provided by TAC compared to standard microbiology cultures.

Secondary objectives included a description of the micro-organisms detected by TAC but not by microbiology culture as well as the impact of TAC on antimicrobial decision-making.

Diagnostic aid and outcomes

A total of 100 children with a median age of 1.2 years (58% male) were included in the study, of whom 80 had suspected community acquired pneumonia and the remainder had hospital acquired pneumonia.

Bacteria were detected more frequently on TAC compared to microbiology cultures (57% vs 18%, p < 0.001)) and In addition, TAC also identified more fungi (17% vs 2%, p < 0.001).

For the detection of bacterial and fungal species, TAC had a sensitivity of 89.5% (95% CI 66.9 – 98.7) and a specificity of 97.9% (95% CI 97.2 – 98.5).

The median time to obtain a result for the diagnostic aid was 25.8 hours compared to 110.4 hours for microbiological cultures and overall, TAC was significantly quicker for both positive and negative results (p < 0.001).

Finally, consultants reported a change of prescription in 47% of cases based upon TAC results. Antimicrobial therapy duration was reduced or stopped in 26% of children, extended in 16% and the spectrum of treatment was broadened in 17% of cases and reduced in 17%.

The authors concluded that as a diagnostic aid, TAC can be used to reliably detect pathogens quicker than routine culture in critically ill children with suspected lower respiratory tract infections. They also called for future studies to incorporate antimicrobial decision support and economic analysis.

Citation
Clark JA et al. The rapid detection of respiratory pathogens in critically ill children. Crit Care, 2023.