Classification of Lung Disease in Children by Using Lung Ultrasound Images and Deep Convolutional Neural Network

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.

Machine learning-aided detection of heart failure (LVEF ≤ 49%) by using ballistocardiography and respiratory effort signals

“…Similar to ( Magrelli et al., 2021 ), the performance metrics include accuracy (Acc), sensitivity (Sen), specificity (Spe), F1 score (F1) and area under the curve (AUC), defined as where TP represents the number of correctly predicted positive samples, TN indicates the number of correctly predicted negative samples, FP is the number of negative samples predicted to be positive, and FN denotes the number of positive samples predicted to be negative. AUC is the area of receiver operating characteristic curve (ROC).…”

“…Similar to (Magrelli et al, 2021), the performance metrics include accuracy (Acc), sensitivity (Sen), specificity (Spe), F1 score (F1) and area under the curve (AUC), defined as…”

An Interpretable Neonatal Lung Ultrasound Feature Extraction and Lung Sliding Detection System Using Object Detectors

“…They utilized transfer learning during their training process. Magrelli et al [17] utilizes Gradient weighted class activation mapping (Grad-CAM) to add interpretability to Image classification models. VGG-19, Xception, Inception-v3, and Inception-ResNet-v2 were trained.…”

“…In our scenario, this approach was not sufficient to detect other LUS features. If we used those feature maps, as described in paper [17] , the colours will overlap, hence, making the interpretation more difficult to read. Physicians require detections to be precise.…”

An Interpretable Object Detection-Based Model for the Diagnosis of Neonatal Lung Diseases Using Ultrasound Images

“…For instance, in RDS we would like to highlight coalescent B-lines with thickened pleura and consolidations. If we utilize those feature maps, as in paper [46] the colors would overlap as shown in figure 2.14 where a Bronchitis scan shows thickened pleura and consolidation. Such model would be difficult for non-skilled users in the NICU to interpret.…”