The global COVID-19 pandemic that started in 2019 and created major disruptions around the world demonstrated the imperative need for quick, inexpensive, accessible and reliable diagnostic methods that would allow the detection of infected individuals with minimal resources. Radiography, and more specifically, chest radiography, is a relatively inexpensive medical imaging modality that can potentially offer a solution for the diagnosis of COVID-19 cases. In this work, we examined eleven deep convolutional neural network architectures for the task of classifying chest X-ray images as belonging to healthy individuals, individuals with COVID-19 or individuals with viral pneumonia. All the examined networks are established architectures that have been proven to be efficient in image classification tasks, and we evaluated three different adjustments to modify the architectures for the task at hand by expanding them with additional layers. The proposed approaches were evaluated for all the examined architectures on a dataset with real chest X-ray images, reaching the highest classification accuracy of 98.04% and the highest F1-score of 98.22% for the best-performing setting.
Lung ultrasound is a widely used portable, cheap, and non-invasive medical imaging technology that can be used to identify various lung pathologies. In this work, we propose a multi-modal approach for lung ultrasound image classification that combines image-based features with information about the type of ultrasound probe used to acquire the input image. Experiments on a large lung ultrasound image dataset that contains images acquired with a linear or a convex ultrasound probe demonstrated the superiority of the proposed approach for the task of classifying lung ultrasound images as "COVID-19", "Normal", "Pneumonia", or "Other", when compared to simply using image-based features. Classification accuracy reached 99.98% using the proposed combination of the Xception pretrained CNN model with the ultrasound probe information, as opposed to 96.81% when only the pre-trained EfficientNetB4 CNN model was used. Furthermore, the experimental results demonstrated a consistent improvement in classification performance when combining the examined base CNN models with probe information, indicating the efficiency of the proposed approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.