High levels of inherent variability in microbiological assessment of bronchoalveolar lavage samples from children with persistent bacterial bronchitis and healthy controls

Time to re-set our thinking about airways disease: lessons from history, the resurgence of chronic bronchitis / PBB and modern concepts in microbiology

In contrast to significant declines in deaths due to lung cancer and cardiac disease in Westernised countries, the mortality due to 'chronic obstructive pulmonary disease' (COPD) has minimally changed in recent decades while 'the incidence of bronchiectasis' is on the rise. The current focus on producing guidelines for these two airway 'diseases' has hindered progress in both treatment and prevention. The elephant in the room is that neither COPD nor bronchiectasis is a disease but rather a consequence of progressive untreated airway inflammation. To make this case, it is important to review the evolution of our understanding of airway disease and how a pathological appearance (bronchiectasis) and an arbitrary physiological marker of impaired airways (COPD) came to be labelled as 'diseases'. Valuable insights into the natural history of airway disease can be obtained from the pre-antibiotic era. The dramatic impacts of antibiotics on the prevalence of significant airway disease, especially in childhood and early adult life, have largely been forgotten and will be revisited as will the misinterpretation of trials undertaken in those with chronic (bacterial) bronchitis. In the past decades, paediatricians have observed a progressive increase in what is termed 'persistent bacterial bronchitis' (PBB). This condition shares all the same characteristics as 'chronic bronchitis', which is prevalent in young children during the pre-antibiotic era. Additionally, the radiological appearance of bronchiectasis is once again becoming more common in children and, more recently, in adults. Adult physicians remain sceptical about the existence of PBB; however, in one study aimed at assessing the efficacy of antibiotics in adults with persistent symptoms, researchers discovered that the majority of patients exhibiting symptoms of PBB were already on long-term macrolides. In recent decades, there has been a growing recognition of the importance of the respiratory microbiome and an understanding of the ability of bacteria to persist in potentially hostile environments through strategies such as biofilms, intracellular communities, and persister bacteria. This is a challenging field that will likely require new approaches to diagnosis and treatment; however, it needs to be embraced if real progress is to be made.

Automated interpretation and analysis of bronchoalveolar lavage fluid

BACKGROUND

The cytological analysis of bronchoalveolar lavage fluid (BALF) plays an essential role in the differential diagnosis of respiratory diseases. In recent years, deep learning has demonstrated excellent performance in image processing and object recognition.

OBJECTIVES

We aim to apply deep learning to the automated interpretation and analysis of BALF.

METHOD

Visual images were acquired using an automated biological microscopy platform. We propose a three-step algorithm to automatically interpret BALF cytology based on a convolutional neural network (CNN). The clinical value was evaluated at the patient level.

RESULTS

Our model successfully detected most cells in BALF specimens and achieved a sensitivity, precision, and F1 score of over 0.9 for most cell types. In two tests in the clinical context, the algorithm outperformed experienced practitioners.

CONCLUSION

The program can automatically provide the cytological background of BALF and augment clinical decision-making for clinicians.