Abstract
Bronchiolitis is the most common cause of hospitalization of children in the first year of life and pneumonia is the leading cause of infant mortality worldwide. Lung ultrasound technology (LUS) is a novel imaging diagnostic tool for the early detection of respiratory distress and offers several advantages due to its low-cost, relative safety, portability, and easy repeatability. More precise and efficient diagnostic and therapeutic strategies are needed. Deep-learning-based computer-aided diagnosis (CADx) systems, using chest X-ray images, have recently demonstrated their potential as a screening tool for pulmonary disease (such as COVID-19 pneumonia). We present the first computer-aided diagnostic scheme for LUS images of pulmonary diseases in children. In this study, we trained from scratch four state-of-the-art deep-learning models (VGG19, Xception, Inception-v3 and Inception-ResNet-v2) for detecting children with bronchiolitis and pneumonia. In our experiments we used a data set consisting of 5,907 images from 33 healthy infants, 3,286 images from 22 infants with bronchiolitis, and 4,769 images from 7 children suffering from bacterial pneumonia. Using four-fold cross-validation, we implemented one binary classification (healthy vs. bronchiolitis) and one three-class classification (healthy vs. bronchiolitis vs. bacterial pneumonia) out of three classes. Affine transformations were applied for data augmentation. Hyperparameters were optimized for the learning rate, dropout regularization, batch size, and epoch iteration. The Inception-ResNet-v2 model provides the highest classification performance, when compared with the other models used on test sets: for healthy vs. bronchiolitis, it provides 97.75% accuracy, 97.75% sensitivity, and 97% specificity whereas for healthy vs. bronchiolitis vs. bacterial pneumonia, the Inception-v3 model provides the best results with 91.5% accuracy, 91.5% sensitivity, and 95.86% specificity. We performed a gradient-weighted class activation mapping (Grad-CAM) visualization and the results were qualitatively evaluated by a pediatrician expert in LUS imaging: heatmaps highlight areas containing diagnostic-relevant LUS imaging-artifacts, e.g., A-, B-, pleural-lines, and consolidations. These complex patterns are automatically learnt from the data, thus avoiding hand-crafted features usage. By using LUS imaging, the proposed framework might aid in the development of an accessible and rapid decision support-method for diagnosing pulmonary diseases in children using LUS imaging.
Highlights
Bronchiolitis is a viral acute lower respiratory-tract infection and the most common reason for hospitalization and intensivecare-unit admission of children worldwide (Choi and Lee, 2012; Øymar et al, 2014).The diagnosis of infants with bronchiolitis is difficult: there exists no unambiguous definition of the disease; the diagnosis is based on clinical evaluation and anamnesis (Ralston et al, 2014) determined by different conditions such as age and variability in the disease state
For experiments classifying healthy vs bronchiolitis vs bacterial pneumonia, we found that the Inception-v3 model provided the best results with a sensitivity of 91.5%, a precision of 92.5%, and an accuracy of 91.5%
EXplainable Artificial Intelligence (XAI) is a newly emerging discipline of AI (Doran et al, 2017) that seeks to develop a series of machine learning (ML) techniques that enable non-expert audiences to better understand and manage results obtained by artificial intelligence (Holzinger et al, 2017)
Summary
The diagnosis of infants with bronchiolitis is difficult: there exists no unambiguous definition of the disease; the diagnosis is based on clinical evaluation and anamnesis (Ralston et al, 2014) determined by different conditions such as age and variability in the disease state. Many of these parameters are based on subjective clinical findings and can be diversely interpreted by different physicians, according to their clinical experience. A diagnosis of CAP, to that of bronchiolitis, relies mainly on medical history and clinical examination. These methods suffer from poor sensitivity and specificity to confirm CAP, physicians need to prescribe medical imaging techniques such as chest X-ray (Bradley et al, 2011; World Health Organization, 2014; Shah et al, 2017)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
